Release v2.3.75

Release v2.3.74
Release v2.3.73
2026-11-08 17:31:00 -10:00 · 2025-12-22 19:11:32 +00:00 · 2025-12-22 17:01:50 +00:00 · 2025-12-22 15:47:51 +00:00 · 2025-12-22 14:50:37 +00:00 · 2025-12-22 13:44:48 +00:00
170 changed files with 35890 additions and 496 deletions
@@ -77,4 +77,7 @@ agents/praisonaiagents/praisonaiagents.egg-info
 .qodo
 CopilotKit*
 .gemini
-issue_details.json
+issue_details.json
+
+# MCP Registry tokens
+.mcpregistry_*
@@ -6,10 +6,13 @@
  </picture>
 </p>

+<!-- mcp-name: io.github.MervinPraison/praisonai -->
+
 <p align="center">
 <a href="https://github.com/MervinPraison/PraisonAI"><img src="https://static.pepy.tech/badge/PraisonAI" alt="Total Downloads" /></a>
 <a href="https://github.com/MervinPraison/PraisonAI"><img src="https://img.shields.io/github/v/release/MervinPraison/PraisonAI" alt="Latest Stable Version" /></a>
 <a href="https://github.com/MervinPraison/PraisonAI"><img src="https://img.shields.io/badge/License-MIT-yellow.svg" alt="License" /></a>
+<a href="https://registry.modelcontextprotocol.io/servers/io.github.MervinPraison/praisonai"><img src="https://img.shields.io/badge/MCP-Registry-blue" alt="MCP Registry" /></a>
 </p>

 <div align="center">
@@ -168,12 +171,13 @@ npm install praisonai

 | Feature | Code | Docs |
 |---------|:----:|:----:|
-| 🚀 Single Agent | [Example](examples/python/agents/single-agent.py) | [📖](https://docs.praison.ai/agents/single) |
-| 🤝 Multi Agents | [Example](examples/python/general/mini_agents_example.py) | [📖](https://docs.praison.ai/concepts/agents) |
-| 🤖 Auto Agents | [Example](examples/python/general/auto_agents_example.py) | [📖](https://docs.praison.ai/features/autoagents) |
-| 🔄 Self Reflection AI Agents | [Example](examples/python/concepts/self-reflection-details.py) | [📖](https://docs.praison.ai/features/selfreflection) |
-| 🧠 Reasoning AI Agents | [Example](examples/python/concepts/reasoning-extraction.py) | [📖](https://docs.praison.ai/features/reasoning) |
-| 👁️ Multi Modal AI Agents | [Example](examples/python/general/multimodal.py) | [📖](https://docs.praison.ai/features/multimodal) |
+| **🤖 Core Agents** | | |
+| ↳ Single Agent | [Example](examples/python/agents/single-agent.py) | [📖](https://docs.praison.ai/agents/single) |
+| ↳ Multi Agents | [Example](examples/python/general/mini_agents_example.py) | [📖](https://docs.praison.ai/concepts/agents) |
+| ↳ Auto Agents | [Example](examples/python/general/auto_agents_example.py) | [📖](https://docs.praison.ai/features/autoagents) |
+| ↳ Self Reflection AI Agents | [Example](examples/python/concepts/self-reflection-details.py) | [📖](https://docs.praison.ai/features/selfreflection) |
+| ↳ Reasoning AI Agents | [Example](examples/python/concepts/reasoning-extraction.py) | [📖](https://docs.praison.ai/features/reasoning) |
+| ↳ Multi Modal AI Agents | [Example](examples/python/general/multimodal.py) | [📖](https://docs.praison.ai/features/multimodal) |
 | **🔄 Workflows** | | |
 | ↳ Simple Workflow | [Example](examples/python/workflows/simple_workflow.py) | [📖](https://docs.praison.ai/features/workflows) |
 | ↳ Workflow with Agents | [Example](examples/python/workflows/workflow_with_agents.py) | [📖](https://docs.praison.ai/features/workflows) |
@@ -185,69 +189,105 @@ npm install praisonai
 | ↳ Workflow Branching | [Example](examples/python/workflows/workflow_branching.py) | [📖](https://docs.praison.ai/features/workflows) |
 | ↳ Workflow Early Stop | [Example](examples/python/workflows/workflow_early_stop.py) | [📖](https://docs.praison.ai/features/workflows) |
 | ↳ Workflow Checkpoints | [Example](examples/python/workflows/workflow_checkpoints.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 📚 Add Custom Knowledge | [Example](examples/python/concepts/knowledge-agents.py) | [📖](https://docs.praison.ai/features/knowledge) |
-| 🧠 Memory (Short & Long Term) | [Example](examples/python/general/memory_example.py) | [📖](https://docs.praison.ai/concepts/memory) |
-| 📄 Chat with PDF Agents | [Example](examples/python/concepts/chat-with-pdf.py) | [📖](https://docs.praison.ai/features/chat-with-pdf) |
-| 💻 Code Interpreter Agents | [Example](examples/python/agents/code-agent.py) | [📖](https://docs.praison.ai/features/codeagent) |
-| ✏️ AI Code Editing Tools | [Example](#ai-code-editing) | [📖](https://docs.praison.ai/features/code) |
-| 📚 RAG Agents | [Example](examples/python/concepts/rag-agents.py) | [📖](https://docs.praison.ai/features/rag) |
-| 🤔 Async & Parallel Processing | [Example](examples/python/general/async_example.py) | [📖](https://docs.praison.ai/features/async) |
-| 🔢 Math Agents | [Example](examples/python/agents/math-agent.py) | [📖](https://docs.praison.ai/features/mathagent) |
-| 🎯 Structured Output Agents | [Example](examples/python/general/structured_agents_example.py) | [📖](https://docs.praison.ai/features/structured) |
-| 🔗 LangChain Integrated Agents | [Example](examples/python/general/langchain_example.py) | [📖](https://docs.praison.ai/features/langchain) |
-| 📞 Callback Agents | [Example](examples/python/general/advanced-callback-systems.py) | [📖](https://docs.praison.ai/features/callbacks) |
-| 🛠️ 100+ Custom Tools | [Example](examples/python/general/tools_example.py) | [📖](https://docs.praison.ai/tools/tools) |
-| 📄 YAML Configuration | [Example](examples/cookbooks/yaml/secondary_market_research_agents.yaml) | [📖](https://docs.praison.ai/developers/agents-playbook) |
-| 💯 100+ LLM Support | [Example](examples/python/providers/openai/openai_gpt4_example.py) | [📖](https://docs.praison.ai/models) |
-| 🔬 Deep Research Agents | [Example](examples/python/agents/research-agent.py) | [📖](https://docs.praison.ai/agents/deep-research) |
-| 🔄 Query Rewriter Agent | [Example](#5-query-rewriter-agent) | [📖](https://docs.praison.ai/agents/query-rewriter) |
-| 🌐 Native Web Search | [Example](examples/python/agents/websearch-agent.py) | [📖](https://docs.praison.ai/agents/websearch) |
-| 📥 Web Fetch (Anthropic) | [Example](#web-search-web-fetch--prompt-caching) | [📖](https://docs.praison.ai/features/model-capabilities) |
-| 💾 Prompt Caching | [Example](#web-search-web-fetch--prompt-caching) | [📖](https://docs.praison.ai/features/model-capabilities) |
-| 🧠 Claude Memory Tool | [Example](#claude-memory-tool-cli) | [📖](https://docs.praison.ai/features/claude-memory-tool) |
-| 💾 File-Based Memory | [Example](examples/python/general/memory_example.py) | [📖](https://docs.praison.ai/concepts/memory) |
-| 🔍 Built-in Search Tools | [Example](examples/python/agents/websearch-agent.py) | [📖](https://docs.praison.ai/tools/tavily) |
-| 📋 Planning Mode | [Example](examples/python/agents/planning-agent.py) | [📖](https://docs.praison.ai/features/planning-mode) |
-| 🔧 Planning Tools | [Example](#3-agent-with-planning-mode) | [📖](https://docs.praison.ai/features/planning-mode) |
-| 🧠 Planning Reasoning | [Example](#3-agent-with-planning-mode) | [📖](https://docs.praison.ai/features/planning-mode) |
-| 🔌 MCP Transports | [Example](examples/python/mcp/mcp-transports-overview.py) | [📖](https://docs.praison.ai/mcp/transports) |
-| 🌐 WebSocket MCP | [Example](examples/python/mcp/websocket-mcp.py) | [📖](https://docs.praison.ai/mcp/sse-transport) |
-| 🔐 MCP Security | [Example](examples/python/mcp/mcp-security.py) | [📖](https://docs.praison.ai/mcp/transports) |
-| 🔄 MCP Resumability | [Example](examples/python/mcp/mcp-resumability.py) | [📖](https://docs.praison.ai/mcp/sse-transport) |
-| ⚡ Fast Context | [Example](examples/context/00_agent_fast_context_basic.py) | [📖](https://docs.praison.ai/features/fast-context) |
-| 🖼️ Image Generation Agent | [Example](examples/python/image/image-agent.py) | [📖](https://docs.praison.ai/features/image-generation) |
-| 📷 Image to Text Agent | [Example](examples/python/agents/image-to-text-agent.py) | [📖](https://docs.praison.ai/agents/image-to-text) |
-| 🎬 Video Agent | [Example](examples/python/agents/video-agent.py) | [📖](https://docs.praison.ai/agents/video) |
-| 📊 Data Analyst Agent | [Example](examples/python/agents/data-analyst-agent.py) | [📖](https://docs.praison.ai/agents/data-analyst) |
-| 💰 Finance Agent | [Example](examples/python/agents/finance-agent.py) | [📖](https://docs.praison.ai/agents/finance) |
-| 🛒 Shopping Agent | [Example](examples/python/agents/shopping-agent.py) | [📖](https://docs.praison.ai/agents/shopping) |
-| ⭐ Recommendation Agent | [Example](examples/python/agents/recommendation-agent.py) | [📖](https://docs.praison.ai/agents/recommendation) |
-| 📖 Wikipedia Agent | [Example](examples/python/agents/wikipedia-agent.py) | [📖](https://docs.praison.ai/agents/wikipedia) |
-| 💻 Programming Agent | [Example](examples/python/agents/programming-agent.py) | [📖](https://docs.praison.ai/agents/programming) |
-| 📝 Markdown Agent | [Example](examples/python/agents/markdown-agent.py) | [📖](https://docs.praison.ai/agents/markdown) |
-| 📝 Prompt Expander Agent | [Example](#prompt-expansion) | [📖](https://docs.praison.ai/agents/prompt-expander) |
-| 🔀 Router Agent | [Example](examples/python/agents/router-agent-cost-optimization.py) | [📖](https://docs.praison.ai/features/routing) |
-| ⛓️ Prompt Chaining | [Example](examples/python/general/prompt_chaining.py) | [📖](https://docs.praison.ai/features/promptchaining) |
-| 🔍 Evaluator Optimiser | [Example](examples/python/general/evaluator-optimiser.py) | [📖](https://docs.praison.ai/features/evaluator-optimiser) |
-| 👷 Orchestrator Workers | [Example](examples/python/general/orchestrator-workers.py) | [📖](https://docs.praison.ai/features/orchestrator-worker) |
-| ⚡ Parallelisation | [Example](examples/python/general/parallelisation.py) | [📖](https://docs.praison.ai/features/parallelisation) |
-| 🔁 Repetitive Agents | [Example](examples/python/concepts/repetitive-agents.py) | [📖](https://docs.praison.ai/features/repetitive) |
-| 🤝 Agent Handoffs | [Example](examples/python/handoff/handoff_basic.py) | [📖](https://docs.praison.ai/features/handoffs) |
-| 🛡️ Guardrails | [Example](examples/python/guardrails/comprehensive-guardrails-example.py) | [📖](https://docs.praison.ai/features/guardrails) |
-| 💬 Sessions Management | [Example](examples/python/sessions/comprehensive-session-management.py) | [📖](https://docs.praison.ai/features/sessions) |
-| ✅ Human Approval | [Example](examples/python/general/human_approval_example.py) | [📖](https://docs.praison.ai/features/approval) |
-| 🔄 Stateful Agents | [Example](examples/python/stateful/workflow-state-example.py) | [📖](https://docs.praison.ai/features/stateful-agents) |
-| 🤖 Autonomous Workflow | [Example](examples/python/general/autonomous-agent.py) | [📖](https://docs.praison.ai/features/autonomous-workflow) |
-| 📜 Rules & Instructions | [Example](#6-rules--instructions) | [📖](https://docs.praison.ai/features/rules) |
-| 🪝 Hooks | [Example](#9-hooks) | [📖](https://docs.praison.ai/features/hooks) |
-| 📈 Telemetry | [Example](examples/python/telemetry/production-telemetry-example.py) | [📖](https://docs.praison.ai/features/telemetry) |
-| 📹 Camera Integration | [Example](examples/python/camera/) | [📖](https://docs.praison.ai/features/camera-integration) |
-| 📄 Project Docs (.praison/docs/) | [Example](#docs-cli) | [📖](https://docs.praison.ai/cli/docs) |
-| 🔌 MCP Config Management | [Example](#mcp-config-cli) | [📖](https://docs.praison.ai/cli/mcp) |
-| 💬 AI Commit Messages | [Example](#ai-commit-cli) | [📖](https://docs.praison.ai/cli/commit) |
-| @ @Mentions in Prompts | [Example](#mentions-in-prompts) | [📖](https://docs.praison.ai/cli/mentions) |
-| 💾 Auto-Save Sessions | [Example](#session-management-python) | [📖](https://docs.praison.ai/cli/session) |
-| 📜 History in Context | [Example](#session-management-python) | [📖](https://docs.praison.ai/cli/session) |
+| **💻 Code & Development** | | |
+| ↳ Code Interpreter Agents | [Example](examples/python/agents/code-agent.py) | [📖](https://docs.praison.ai/features/codeagent) |
+| ↳ AI Code Editing Tools | [Example](examples/python/code/code_editing_example.py) | [📖](https://docs.praison.ai/code/editing) |
+| ↳ External Agents (All) | [Example](examples/python/code/external_agents_example.py) | [📖](https://docs.praison.ai/code/external-agents) |
+| ↳ Claude Code CLI | [Example](examples/python/code/claude_code_example.py) | [📖](https://docs.praison.ai/code/claude-code) |
+| ↳ Gemini CLI | [Example](examples/python/code/gemini_cli_example.py) | [📖](https://docs.praison.ai/code/gemini-cli) |
+| ↳ Codex CLI | [Example](examples/python/code/codex_cli_example.py) | [📖](https://docs.praison.ai/code/codex-cli) |
+| ↳ Cursor CLI | [Example](examples/python/code/cursor_cli_example.py) | [📖](https://docs.praison.ai/code/cursor-cli) |
+| **🧠 Memory & Knowledge** | | |
+| ↳ Memory (Short & Long Term) | [Example](examples/python/general/memory_example.py) | [📖](https://docs.praison.ai/concepts/memory) |
+| ↳ File-Based Memory | [Example](examples/python/general/memory_example.py) | [📖](https://docs.praison.ai/concepts/memory) |
+| ↳ Claude Memory Tool | [Example](#claude-memory-tool-cli) | [📖](https://docs.praison.ai/features/claude-memory-tool) |
+| ↳ Add Custom Knowledge | [Example](examples/python/concepts/knowledge-agents.py) | [📖](https://docs.praison.ai/features/knowledge) |
+| ↳ RAG Agents | [Example](examples/python/concepts/rag-agents.py) | [📖](https://docs.praison.ai/features/rag) |
+| ↳ Chat with PDF Agents | [Example](examples/python/concepts/chat-with-pdf.py) | [📖](https://docs.praison.ai/features/chat-with-pdf) |
+| **🔬 Research & Intelligence** | | |
+| ↳ Deep Research Agents | [Example](examples/python/agents/research-agent.py) | [📖](https://docs.praison.ai/agents/deep-research) |
+| ↳ Query Rewriter Agent | [Example](#5-query-rewriter-agent) | [📖](https://docs.praison.ai/agents/query-rewriter) |
+| ↳ Native Web Search | [Example](examples/python/agents/websearch-agent.py) | [📖](https://docs.praison.ai/agents/websearch) |
+| ↳ Built-in Search Tools | [Example](examples/python/agents/websearch-agent.py) | [📖](https://docs.praison.ai/tools/tavily) |
+| ↳ Unified Web Search | [Example](src/praisonai-agents/examples/web_search_example.py) | [📖](https://docs.praison.ai/tools/web-search) |
+| ↳ Web Fetch (Anthropic) | [Example](#web-search-web-fetch--prompt-caching) | [📖](https://docs.praison.ai/features/model-capabilities) |
+| **📋 Planning & Execution** | | |
+| ↳ Planning Mode | [Example](examples/python/agents/planning-agent.py) | [📖](https://docs.praison.ai/features/planning-mode) |
+| ↳ Planning Tools | [Example](#3-agent-with-planning-mode) | [📖](https://docs.praison.ai/features/planning-mode) |
+| ↳ Planning Reasoning | [Example](#3-agent-with-planning-mode) | [📖](https://docs.praison.ai/features/planning-mode) |
+| ↳ Prompt Chaining | [Example](examples/python/general/prompt_chaining.py) | [📖](https://docs.praison.ai/features/promptchaining) |
+| ↳ Evaluator Optimiser | [Example](examples/python/general/evaluator-optimiser.py) | [📖](https://docs.praison.ai/features/evaluator-optimiser) |
+| ↳ Orchestrator Workers | [Example](examples/python/general/orchestrator-workers.py) | [📖](https://docs.praison.ai/features/orchestrator-worker) |
+| **👥 Specialized Agents** | | |
+| ↳ Data Analyst Agent | [Example](examples/python/agents/data-analyst-agent.py) | [📖](https://docs.praison.ai/agents/data-analyst) |
+| ↳ Finance Agent | [Example](examples/python/agents/finance-agent.py) | [📖](https://docs.praison.ai/agents/finance) |
+| ↳ Shopping Agent | [Example](examples/python/agents/shopping-agent.py) | [📖](https://docs.praison.ai/agents/shopping) |
+| ↳ Recommendation Agent | [Example](examples/python/agents/recommendation-agent.py) | [📖](https://docs.praison.ai/agents/recommendation) |
+| ↳ Wikipedia Agent | [Example](examples/python/agents/wikipedia-agent.py) | [📖](https://docs.praison.ai/agents/wikipedia) |
+| ↳ Programming Agent | [Example](examples/python/agents/programming-agent.py) | [📖](https://docs.praison.ai/agents/programming) |
+| ↳ Math Agents | [Example](examples/python/agents/math-agent.py) | [📖](https://docs.praison.ai/features/mathagent) |
+| ↳ Markdown Agent | [Example](examples/python/agents/markdown-agent.py) | [📖](https://docs.praison.ai/agents/markdown) |
+| ↳ Prompt Expander Agent | [Example](#prompt-expansion) | [📖](https://docs.praison.ai/agents/prompt-expander) |
+| **🎨 Media & Multimodal** | | |
+| ↳ Image Generation Agent | [Example](examples/python/image/image-agent.py) | [📖](https://docs.praison.ai/features/image-generation) |
+| ↳ Image to Text Agent | [Example](examples/python/agents/image-to-text-agent.py) | [📖](https://docs.praison.ai/agents/image-to-text) |
+| ↳ Video Agent | [Example](examples/python/agents/video-agent.py) | [📖](https://docs.praison.ai/agents/video) |
+| ↳ Camera Integration | [Example](examples/python/camera/) | [📖](https://docs.praison.ai/features/camera-integration) |
+| **🔌 Protocols & Integration** | | |
+| ↳ MCP Transports | [Example](examples/python/mcp/mcp-transports-overview.py) | [📖](https://docs.praison.ai/mcp/transports) |
+| ↳ WebSocket MCP | [Example](examples/python/mcp/websocket-mcp.py) | [📖](https://docs.praison.ai/mcp/sse-transport) |
+| ↳ MCP Security | [Example](examples/python/mcp/mcp-security.py) | [📖](https://docs.praison.ai/mcp/transports) |
+| ↳ MCP Resumability | [Example](examples/python/mcp/mcp-resumability.py) | [📖](https://docs.praison.ai/mcp/sse-transport) |
+| ↳ MCP Config Management | [Example](#mcp-config-cli) | [📖](https://docs.praison.ai/docs/cli/mcp) |
+| ↳ LangChain Integrated Agents | [Example](examples/python/general/langchain_example.py) | [📖](https://docs.praison.ai/features/langchain) |
+| **🛡️ Safety & Control** | | |
+| ↳ Guardrails | [Example](examples/python/guardrails/comprehensive-guardrails-example.py) | [📖](https://docs.praison.ai/features/guardrails) |
+| ↳ Human Approval | [Example](examples/python/general/human_approval_example.py) | [📖](https://docs.praison.ai/features/approval) |
+| ↳ Rules & Instructions | [Example](#6-rules--instructions) | [📖](https://docs.praison.ai/features/rules) |
+| **⚙️ Advanced Features** | | |
+| ↳ Async & Parallel Processing | [Example](examples/python/general/async_example.py) | [📖](https://docs.praison.ai/features/async) |
+| ↳ Parallelisation | [Example](examples/python/general/parallelisation.py) | [📖](https://docs.praison.ai/features/parallelisation) |
+| ↳ Repetitive Agents | [Example](examples/python/concepts/repetitive-agents.py) | [📖](https://docs.praison.ai/features/repetitive) |
+| ↳ Agent Handoffs | [Example](examples/python/handoff/handoff_basic.py) | [📖](https://docs.praison.ai/features/handoffs) |
+| ↳ Stateful Agents | [Example](examples/python/stateful/workflow-state-example.py) | [📖](https://docs.praison.ai/features/stateful-agents) |
+| ↳ Autonomous Workflow | [Example](examples/python/general/autonomous-agent.py) | [📖](https://docs.praison.ai/features/autonomous-workflow) |
+| ↳ Structured Output Agents | [Example](examples/python/general/structured_agents_example.py) | [📖](https://docs.praison.ai/features/structured) |
+| ↳ Model Router | [Example](examples/python/agents/router-agent-cost-optimization.py) | [📖](https://docs.praison.ai/features/model-router) |
+| ↳ Prompt Caching | [Example](#web-search-web-fetch--prompt-caching) | [📖](https://docs.praison.ai/features/model-capabilities) |
+| ↳ Fast Context | [Example](examples/context/00_agent_fast_context_basic.py) | [📖](https://docs.praison.ai/features/fast-context) |
+| **🛠️ Tools & Configuration** | | |
+| ↳ 100+ Custom Tools | [Example](examples/python/general/tools_example.py) | [📖](https://docs.praison.ai/tools/tools) |
+| ↳ YAML Configuration | [Example](examples/cookbooks/yaml/secondary_market_research_agents.yaml) | [📖](https://docs.praison.ai/developers/agents-playbook) |
+| ↳ 100+ LLM Support | [Example](examples/python/providers/openai/openai_gpt4_example.py) | [📖](https://docs.praison.ai/models) |
+| ↳ Callback Agents | [Example](examples/python/general/advanced-callback-systems.py) | [📖](https://docs.praison.ai/features/callbacks) |
+| ↳ Hooks | [Example](#9-hooks) | [📖](https://docs.praison.ai/features/hooks) |
+| **📊 Monitoring & Management** | | |
+| ↳ Sessions Management | [Example](examples/python/sessions/comprehensive-session-management.py) | [📖](https://docs.praison.ai/features/sessions) |
+| ↳ Auto-Save Sessions | [Example](#session-management-python) | [📖](https://docs.praison.ai/docs/cli/session) |
+| ↳ History in Context | [Example](#session-management-python) | [📖](https://docs.praison.ai/docs/cli/session) |
+| ↳ Telemetry | [Example](examples/python/telemetry/production-telemetry-example.py) | [📖](https://docs.praison.ai/features/telemetry) |
+| ↳ Project Docs (.praison/docs/) | [Example](#docs-cli) | [📖](https://docs.praison.ai/docs/cli/docs) |
+| ↳ AI Commit Messages | [Example](#ai-commit-cli) | [📖](https://docs.praison.ai/docs/cli/commit) |
+| ↳ @Mentions in Prompts | [Example](#mentions-in-prompts) | [📖](https://docs.praison.ai/docs/cli/mentions) |
+| **🖥️ CLI Features** | | |
+| ↳ Slash Commands | [Example](examples/python/cli/slash_commands_example.py) | [📖](https://docs.praison.ai/docs/cli/slash-commands) |
+| ↳ Autonomy Modes | [Example](examples/python/cli/autonomy_modes_example.py) | [📖](https://docs.praison.ai/docs/cli/autonomy-modes) |
+| ↳ Cost Tracking | [Example](examples/python/cli/cost_tracking_example.py) | [📖](https://docs.praison.ai/docs/cli/cost-tracking) |
+| ↳ Repository Map | [Example](examples/python/cli/repo_map_example.py) | [📖](https://docs.praison.ai/docs/cli/repo-map) |
+| ↳ Interactive TUI | [Example](examples/python/cli/interactive_tui_example.py) | [📖](https://docs.praison.ai/docs/cli/interactive-tui) |
+| ↳ Git Integration | [Example](examples/python/cli/git_integration_example.py) | [📖](https://docs.praison.ai/docs/cli/git-integration) |
+| ↳ Sandbox Execution | [Example](examples/python/cli/sandbox_execution_example.py) | [📖](https://docs.praison.ai/docs/cli/sandbox-execution) |
+| ↳ CLI Compare | [Example](examples/compare/cli_compare_basic.py) | [📖](https://docs.praison.ai/docs/cli/compare) |
+| **🧪 Evaluation** | | |
+| ↳ Accuracy Evaluation | [Example](examples/eval/accuracy_example.py) | [📖](https://docs.praison.ai/docs/cli/eval) |
+| ↳ Performance Evaluation | [Example](examples/eval/performance_example.py) | [📖](https://docs.praison.ai/docs/cli/eval) |
+| ↳ Reliability Evaluation | [Example](examples/eval/reliability_example.py) | [📖](https://docs.praison.ai/docs/cli/eval) |
+| ↳ Criteria Evaluation | [Example](examples/eval/criteria_example.py) | [📖](https://docs.praison.ai/docs/cli/eval) |
+| **🎯 Agent Skills** | | |
+| ↳ Skills Management | [Example](examples/skills/basic_skill_usage.py) | [📖](https://docs.praison.ai/features/skills) |
+| ↳ Custom Skills | [Example](examples/skills/custom_skill_example.py) | [📖](https://docs.praison.ai/features/skills) |
+| **⏰ 24/7 Scheduling** | | |
+| ↳ Agent Scheduler | [Example](examples/python/scheduled_agents/news_checker_live.py) | [📖](https://docs.praison.ai/docs/cli/scheduler) |

 ---

@@ -954,6 +994,32 @@ export OPENAI_API_KEY=xxxxxxxxxxxxxxxxxxxxxx
 praisonai --auto create a movie script about Robots in Mars
 ```

+### Interactive Mode CLI:
+```bash
+# Start interactive terminal mode (inspired by Gemini CLI, Codex CLI, Claude Code)
+praisonai --interactive
+praisonai -i
+
+# Features:
+# - Streaming responses (no boxes)
+# - Built-in tools: read_file, write_file, list_files, execute_command, internet_search
+# - Slash commands: /help, /exit, /tools, /clear
+
+# Chat mode - single prompt with interactive style (for testing/scripting)
+# Use --chat (or --chat-mode for backward compatibility)
+praisonai "list files in current folder" --chat
+praisonai "search the web for AI news" --chat
+praisonai "read README.md" --chat
+```
+
+### Chat UI (Web Interface):
+```bash
+# Start web-based Chainlit chat interface (requires praisonai[chat])
+pip install "praisonai[chat]"
+praisonai chat
+# Opens browser at http://localhost:8084
+```
+
 ### Query Rewriting (works with any command):
 ```bash
 # Rewrite query for better results (uses QueryRewriterAgent)
@@ -1014,6 +1080,20 @@ praisonai "Complex analysis task" --planning --planning-reasoning
 praisonai "Task" --planning --auto-approve-plan
 ```

+### Tool Approval CLI:
+```bash
+# Auto-approve ALL tool executions (use with caution!)
+praisonai "run ls command" --trust
+
+# Auto-approve tools up to a risk level (prompt for higher)
+# Levels: low, medium, high, critical
+praisonai "write to file" --approve-level high  # Prompts for critical tools only
+praisonai "task" --approve-level medium         # Prompts for high and critical
+
+# Default behavior (no flags): prompts for all dangerous tools
+praisonai "run shell command"  # Will prompt for approval
+```
+
 ### Memory CLI:
 ```bash
 # Enable memory for agent (persists across sessions)
@@ -1120,6 +1200,15 @@ praisonai workflow auto "Research AI trends" --pattern parallel
 # Generate routing workflow (classifier routes to specialists)
 praisonai workflow auto "Build a chatbot" --pattern routing

+# Generate orchestrator-workers workflow (central orchestrator delegates)
+praisonai workflow auto "Comprehensive market analysis" --pattern orchestrator-workers
+
+# Generate evaluator-optimizer workflow (iterative refinement)
+praisonai workflow auto "Write and refine article" --pattern evaluator-optimizer
+
+# Specify output file
+praisonai workflow auto "Build a chatbot" --pattern routing
+
 # Specify output file
 praisonai workflow auto "Research AI" --pattern sequential --output my_workflow.yaml
 ```
@@ -1129,7 +1218,7 @@ praisonai workflow auto "Research AI" --pattern sequential --output my_workflow.
 |------|-------------|
 | `--workflow-var key=value` | Set workflow variable (can be repeated) |
 | `--var key=value` | Set variable for YAML workflows |
-| `--pattern <pattern>` | Pattern for auto-generation (sequential, parallel, routing, loop) |
+| `--pattern <pattern>` | Pattern for auto-generation (sequential, parallel, routing, loop, orchestrator-workers, evaluator-optimizer) |
 | `--output <file>` | Output file for auto-generation |
 | `--llm <model>` | LLM model (e.g., openai/gpt-4o-mini) |
 | `--tools <tools>` | Tools (comma-separated, e.g., tavily) |
@@ -1175,6 +1264,21 @@ praisonai "Analyze this data" --metrics
 praisonai "Complex task" --metrics --planning
 ```

+### Scheduler CLI:
+
+```bash
+praisonai schedule start <name> "task" --interval hourly
+praisonai schedule list
+praisonai schedule logs <name> [--follow]
+praisonai schedule stop <name>
+praisonai schedule restart <name>
+praisonai schedule delete <name>
+praisonai schedule describe <name>
+praisonai schedule save <name> [file.yaml]
+praisonai schedule "task" --interval hourly  # foreground mode
+praisonai schedule agents.yaml  # foreground mode
+```
+
 ### Image Processing CLI:
 ```bash
 # Process images with vision-based tasks
@@ -1361,8 +1465,13 @@ praisonai "Complex analysis" --router --router-provider anthropic
 # Router automatically selects:
 # - Simple tasks → gpt-4o-mini, claude-3-haiku
 # - Complex tasks → gpt-4-turbo, claude-3-opus
+
+# Create workflow with model routing template
+praisonai workflow create --template model-routing --output my_workflow.yaml
 ```

+Custom models can be configured in `agents.yaml`. See [Model Router Docs](https://docs.praison.ai/features/model-router) for details.
+
 ### Flow Display CLI:
 ```bash
 # Enable visual workflow tracking
@@ -1414,13 +1523,31 @@ praisonai mcp help

 ### AI Commit CLI:
 ```bash
-# Generate AI commit message for staged changes
+# Full auto mode: stage all, security check, commit, and push
+praisonai commit -a
+
+# Interactive mode (requires git add first)
 praisonai commit

-# Generate, commit, and push
+# Interactive with auto-push
 praisonai commit --push
+
+# Skip security check (not recommended)
+praisonai commit -a --no-verify
 ```

+**Features:**
+- 🤖 AI-generated conventional commit messages
+- 🔒 Built-in security scanning (API keys, passwords, secrets, sensitive files)
+- 📦 Auto-staging with `-a` flag
+- 🚀 Auto-push in full auto mode
+- ✏️ Edit message before commit in interactive mode
+
+**Security Detection:**
+- API keys, secrets, tokens (AWS, GitHub, GitLab, Slack)
+- Passwords and private keys
+- Sensitive files (`.env`, `id_rsa`, `.pem`, `.key`, etc.)
+
 ### Serve CLI (API Server):
 ```bash
 # Start API server for agents defined in YAML
@@ -1451,6 +1578,55 @@ export N8N_API_KEY="your-api-key"
 praisonai agents.yaml --n8n
 ```

+### External Agents CLI:
+
+Use external AI coding CLI tools (Claude Code, Gemini CLI, Codex CLI, Cursor CLI) as agent tools:
+
+```bash
+# Use Claude Code for coding tasks
+praisonai "Refactor the auth module" --external-agent claude
+
+# Use Gemini CLI for code analysis
+praisonai "Analyze codebase architecture" --external-agent gemini
+
+# Use OpenAI Codex CLI
+praisonai "Fix all bugs in src/" --external-agent codex
+
+# Use Cursor CLI
+praisonai "Add comprehensive tests" --external-agent cursor
+```
+
+**Python API:**
+```python
+from praisonai.integrations import (
+    ClaudeCodeIntegration,
+    GeminiCLIIntegration,
+    CodexCLIIntegration,
+    CursorCLIIntegration
+)
+
+# Create integration
+claude = ClaudeCodeIntegration(workspace="/project")
+
+# Execute a coding task
+result = await claude.execute("Refactor the auth module")
+
+# Use as agent tool
+from praisonai import Agent
+tool = claude.as_tool()
+agent = Agent(tools=[tool])
+```
+
+**Environment Variables:**
+```bash
+export ANTHROPIC_API_KEY=your-key  # Claude Code
+export GEMINI_API_KEY=your-key     # Gemini CLI
+export OPENAI_API_KEY=your-key     # Codex CLI
+export CURSOR_API_KEY=your-key     # Cursor CLI
+```
+
+See [External Agents Documentation](https://docs.praison.ai/code/external-agents) for more details.
+
 ### @Mentions in Prompts:
 ```bash
 # Include file content in prompt
@@ -1676,32 +1852,32 @@ app.include_router(a2a.get_router())

 | Feature | Docs |
 |---------|:----:|
-| 🔄 Query Rewrite - RAG optimization | [📖](https://docs.praison.ai/cli/query-rewrite) |
-| 🔬 Deep Research - Automated research | [📖](https://docs.praison.ai/cli/deep-research) |
-| 📋 Planning - Step-by-step execution | [📖](https://docs.praison.ai/cli/planning) |
-| 💾 Memory - Persistent agent memory | [📖](https://docs.praison.ai/cli/memory) |
-| 📜 Rules - Auto-discovered instructions | [📖](https://docs.praison.ai/cli/rules) |
-| 🔄 Workflow - Multi-step workflows | [📖](https://docs.praison.ai/cli/workflow) |
-| 🪝 Hooks - Event-driven actions | [📖](https://docs.praison.ai/cli/hooks) |
-| 🧠 Claude Memory - Anthropic memory tool | [📖](https://docs.praison.ai/cli/claude-memory) |
-| 🛡️ Guardrail - Output validation | [📖](https://docs.praison.ai/cli/guardrail) |
-| 📊 Metrics - Token usage tracking | [📖](https://docs.praison.ai/cli/metrics) |
-| 🖼️ Image - Vision processing | [📖](https://docs.praison.ai/cli/image) |
-| 📡 Telemetry - Usage monitoring | [📖](https://docs.praison.ai/cli/telemetry) |
-| 🔌 MCP - Model Context Protocol | [📖](https://docs.praison.ai/cli/mcp) |
-| ⚡ Fast Context - Codebase search | [📖](https://docs.praison.ai/cli/fast-context) |
-| 📚 Knowledge - RAG management | [📖](https://docs.praison.ai/cli/knowledge) |
-| 💬 Session - Conversation management | [📖](https://docs.praison.ai/cli/session) |
-| 🔧 Tools - Tool discovery | [📖](https://docs.praison.ai/cli/tools) |
-| 🤝 Handoff - Agent delegation | [📖](https://docs.praison.ai/cli/handoff) |
-| 🧠 Auto Memory - Memory extraction | [📖](https://docs.praison.ai/cli/auto-memory) |
-| 📋 Todo - Task management | [📖](https://docs.praison.ai/cli/todo) |
-| 🎯 Router - Smart model selection | [📖](https://docs.praison.ai/cli/router) |
-| 📈 Flow Display - Visual workflow | [📖](https://docs.praison.ai/cli/flow-display) |
-| ✨ Prompt Expansion - Detailed prompts | [📖](https://docs.praison.ai/cli/prompt-expansion) |
-| 🌐 Web Search - Real-time search | [📖](https://docs.praison.ai/cli/web-search) |
-| 📥 Web Fetch - URL content retrieval | [📖](https://docs.praison.ai/cli/web-fetch) |
-| 💾 Prompt Caching - Cost reduction | [📖](https://docs.praison.ai/cli/prompt-caching) |
+| 🔄 Query Rewrite - RAG optimization | [📖](https://docs.praison.ai/docs/cli/query-rewrite) |
+| 🔬 Deep Research - Automated research | [📖](https://docs.praison.ai/docs/cli/deep-research) |
+| 📋 Planning - Step-by-step execution | [📖](https://docs.praison.ai/docs/cli/planning) |
+| 💾 Memory - Persistent agent memory | [📖](https://docs.praison.ai/docs/cli/memory) |
+| 📜 Rules - Auto-discovered instructions | [📖](https://docs.praison.ai/docs/cli/rules) |
+| 🔄 Workflow - Multi-step workflows | [📖](https://docs.praison.ai/docs/cli/workflow) |
+| 🪝 Hooks - Event-driven actions | [📖](https://docs.praison.ai/docs/cli/hooks) |
+| 🧠 Claude Memory - Anthropic memory tool | [📖](https://docs.praison.ai/docs/cli/claude-memory) |
+| 🛡️ Guardrail - Output validation | [📖](https://docs.praison.ai/docs/cli/guardrail) |
+| 📊 Metrics - Token usage tracking | [📖](https://docs.praison.ai/docs/cli/metrics) |
+| 🖼️ Image - Vision processing | [📖](https://docs.praison.ai/docs/cli/image) |
+| 📡 Telemetry - Usage monitoring | [📖](https://docs.praison.ai/docs/cli/telemetry) |
+| 🔌 MCP - Model Context Protocol | [📖](https://docs.praison.ai/docs/cli/mcp) |
+| ⚡ Fast Context - Codebase search | [📖](https://docs.praison.ai/docs/cli/fast-context) |
+| 📚 Knowledge - RAG management | [📖](https://docs.praison.ai/docs/cli/knowledge) |
+| 💬 Session - Conversation management | [📖](https://docs.praison.ai/docs/cli/session) |
+| 🔧 Tools - Tool discovery | [📖](https://docs.praison.ai/docs/cli/tools) |
+| 🤝 Handoff - Agent delegation | [📖](https://docs.praison.ai/docs/cli/handoff) |
+| 🧠 Auto Memory - Memory extraction | [📖](https://docs.praison.ai/docs/cli/auto-memory) |
+| 📋 Todo - Task management | [📖](https://docs.praison.ai/docs/cli/todo) |
+| 🎯 Router - Smart model selection | [📖](https://docs.praison.ai/docs/cli/router) |
+| 📈 Flow Display - Visual workflow | [📖](https://docs.praison.ai/docs/cli/flow-display) |
+| ✨ Prompt Expansion - Detailed prompts | [📖](https://docs.praison.ai/docs/cli/prompt-expansion) |
+| 🌐 Web Search - Real-time search | [📖](https://docs.praison.ai/docs/cli/web-search) |
+| 📥 Web Fetch - URL content retrieval | [📖](https://docs.praison.ai/docs/cli/web-fetch) |
+| 💾 Prompt Caching - Cost reduction | [📖](https://docs.praison.ai/docs/cli/prompt-caching) |

 ---

@@ -2253,7 +2429,7 @@ PraisonAI provides zero-dependency persistent memory for agents. For detailed ex
 - 📖 **Wikipedia Agent** - Search and extract Wikipedia information
 - 💻 **Programming Agent** - Code development and analysis
 - 📝 **Markdown Agent** - Generate and format Markdown content
- 🔀 **Router Agent** - Dynamic task routing with cost optimisation
+- 🔀 **Model Router** - Smart model selection based on task complexity

 ### MCP Protocol

@@ -2274,6 +2450,7 @@ PraisonAI provides zero-dependency persistent memory for agents. For detailed ex
 - 🤝 **Agent Handoffs** - Transfer context between specialised agents
 - 🛡️ **Guardrails** - Input/output validation and safety checks
 - ✅ **Human Approval** - Require human confirmation for critical actions
+- 🔐 **Tool Approval CLI** - `--trust` (auto-approve all) and `--approve-level` (risk-based approval)
 - 💬 **Sessions Management** - Isolated conversation contexts
 - 🔄 **Stateful Agents** - Maintain state across interactions

@@ -16,7 +16,7 @@ RUN mkdir -p /root/.praison
 # Install Python packages (using latest versions)
 RUN pip install --no-cache-dir \
    praisonai_tools \
-    "praisonai>=2.3.41" \
+    "praisonai>=2.3.75" \
    "praisonai[chat]" \
    "embedchain[github,youtube]"

@@ -20,7 +20,7 @@ RUN mkdir -p /root/.praison
 # Install Python packages (using latest versions)
 RUN pip install --no-cache-dir \
    praisonai_tools \
-    "praisonai>=2.3.41" \
+    "praisonai>=2.3.75" \
    "praisonai[ui]" \
    "praisonai[chat]" \
    "praisonai[realtime]" \
@@ -16,7 +16,7 @@ RUN mkdir -p /root/.praison
 # Install Python packages (using latest versions)
 RUN pip install --no-cache-dir \
    praisonai_tools \
-    "praisonai>=2.3.41" \
+    "praisonai>=2.3.75" \
    "praisonai[ui]" \
    "praisonai[crewai]"

@@ -0,0 +1,49 @@
+# PraisonAI Examples
+
+This folder contains examples for PraisonAI. For detailed documentation, visit [docs.praison.ai](https://docs.praison.ai).
+
+## Structure
+
+```
+examples/
+├── python/           # Python examples
+│   ├── agents/       # Agent examples (single, multi, router, etc.)
+│   ├── workflows/    # Workflow patterns (routing, parallel, loop)
+│   ├── tools/        # Custom tools examples
+│   ├── mcp/          # MCP protocol examples
+│   ├── memory/       # Memory and sessions
+│   ├── code/         # Code editing and external CLI tools
+│   └── ...
+├── yaml/             # YAML workflow examples
+└── cookbooks/        # Complete use-case examples
+```
+
+## Quick Links
+
+| Category | Examples | Docs |
+|----------|----------|------|
+| **Agents** | [python/agents/](python/agents/) | [📖](https://docs.praison.ai/concepts/agents) |
+| **Workflows** | [python/workflows/](python/workflows/) | [📖](https://docs.praison.ai/features/workflows) |
+| **Model Router** | [python/agents/router-agent-cost-optimization.py](python/agents/router-agent-cost-optimization.py) | [📖](https://docs.praison.ai/features/model-router) |
+| **MCP** | [python/mcp/](python/mcp/) | [📖](https://docs.praison.ai/mcp) |
+| **Memory** | [python/memory/](python/memory/) | [📖](https://docs.praison.ai/concepts/memory) |
+| **Tools** | [python/tools/](python/tools/) | [📖](https://docs.praison.ai/tools) |
+| **Code** | [python/code/](python/code/) | [📖](https://docs.praison.ai/code) |
+| **YAML** | [yaml/](yaml/) | [📖](https://docs.praison.ai/features/yaml-workflows) |
+
+## Running Examples
+
+```bash
+# Install PraisonAI
+pip install praisonai
+
+# Set API key
+export OPENAI_API_KEY=your_key_here
+
+# Run an example
+python examples/python/agents/single-agent.py
+```
+
+## CLI Commands
+
+See the main [README.md](../README.md#-cli--no-code-interface) for all CLI commands.
@@ -0,0 +1,85 @@
+"""
+CLI Compare Example - Advanced Usage
+
+This example demonstrates advanced comparison features including
+custom mode configurations and result analysis.
+"""
+
+from praisonai.cli.features.compare import (
+    CompareHandler,
+    CompareResult,
+    ModeResult,
+    get_mode_config,
+    save_compare_result,
+    format_comparison_table,
+)
+
+# Create custom mode configurations
+custom_modes = {
+    "fast": {"llm": "gpt-4o-mini"},
+    "accurate": {"llm": "gpt-4o", "planning": True},
+    "research": {"research": True, "web_search": True},
+}
+
+# Create handler with verbose output
+handler = CompareHandler(verbose=True)
+
+# Example: Manually create comparison results for analysis
+# (useful for testing without API calls)
+mock_result = CompareResult(
+    query="What are the latest AI trends?",
+    comparisons=[
+        ModeResult(
+            mode="basic",
+            output="AI trends include LLMs, multimodal models, and agents.",
+            execution_time_ms=1234.5,
+            model_used="gpt-4o-mini",
+            tools_used=[]
+        ),
+        ModeResult(
+            mode="tools",
+            output="Based on recent search results, AI trends in 2024 include...",
+            execution_time_ms=2567.8,
+            model_used="gpt-4o-mini",
+            tools_used=["internet_search"]
+        ),
+        ModeResult(
+            mode="research",
+            output="Comprehensive research shows that AI trends...",
+            execution_time_ms=5432.1,
+            model_used="gpt-4o-mini",
+            tools_used=["internet_search", "web_scraper"]
+        ),
+    ]
+)
+
+# Get summary statistics
+summary = mock_result.get_summary()
+print("Comparison Summary:")
+print(f"  Fastest mode: {summary['fastest']} ({summary['fastest_time_ms']:.1f}ms)")
+print(f"  Slowest mode: {summary['slowest']} ({summary['slowest_time_ms']:.1f}ms)")
+
+# Format as table
+print("\n" + format_comparison_table(mock_result))
+
+# Convert to JSON for storage
+print("\nJSON Output:")
+print(mock_result.to_json())
+
+# Save to file
+# save_compare_result(mock_result, "advanced_comparison.json")
+
+print("\n--- CLI Usage Examples ---")
+print("""
+# Basic comparison
+praisonai "What is AI?" --compare "basic,tools"
+
+# Compare with specific model
+praisonai "Explain quantum computing" --compare "basic,planning,research" --model gpt-4o
+
+# Save results to file
+praisonai "Latest tech news" --compare "basic,tools,web_search" --compare-output results.json
+
+# Verbose output
+praisonai "Write a poem" --compare "basic,planning" --verbose
+""")
@@ -0,0 +1,45 @@
+"""
+CLI Compare Example - Basic Usage
+
+This example demonstrates how to compare different CLI modes programmatically.
+"""
+
+from praisonai.cli.features.compare import (
+    CompareHandler,
+    get_mode_config,
+    list_available_modes,
+    parse_modes,
+)
+
+# List all available comparison modes
+print("Available modes:")
+for mode in list_available_modes():
+    config = get_mode_config(mode)
+    print(f"  - {mode}: {config}")
+
+# Parse modes from comma-separated string
+modes = parse_modes("basic,tools,planning")
+print(f"\nParsed modes: {modes}")
+
+# Create compare handler
+handler = CompareHandler(verbose=True)
+
+# Compare modes (this will run actual agent calls)
+# Uncomment to run:
+# result = handler.compare(
+#     query="What is artificial intelligence?",
+#     modes=["basic", "tools"],
+#     model="gpt-4o-mini"
+# )
+# handler.print_result(result)
+
+# Save results to file
+# result = handler.execute(
+#     query="Explain machine learning",
+#     modes_str="basic,planning",
+#     output_path="comparison_results.json"
+# )
+
+print("\nTo run comparison via CLI:")
+print('  praisonai "What is AI?" --compare "basic,tools,planning"')
+print('  praisonai "What is AI?" --compare "basic,research" --compare-output results.json')
@@ -0,0 +1,23 @@
+# Example: Data Collection Agent with 30-minute Schedule
+# Collects and analyzes data every 30 minutes
+
+framework: praisonai
+
+agents:
+  - name: "Data Collector"
+    role: "Data Analysis Specialist"
+    goal: "Collect and analyze market data at regular intervals"
+    instructions: |
+      You are a data analysis specialist focused on collecting and analyzing
+      market trends and metrics. Provide actionable insights from the data.
+    tools:
+      - search_tool
+    verbose: false
+
+task: "Collect latest market data and identify key trends"
+
+# Schedule: Every 30 minutes with immediate execution
+schedule:
+  interval: "*/30m"
+  max_retries: 5
+  run_immediately: true
@@ -0,0 +1,53 @@
+# Example: 24/7 AI News Monitor with Scheduler
+# This agent runs continuously at regular intervals to monitor AI news
+
+framework: praisonai
+
+agents:
+  - name: "AI News Monitor"
+    role: "Technology News Analyst"
+    goal: "Monitor and summarize the latest AI developments"
+    instructions: |
+      You are an expert technology news analyst specializing in AI and machine learning.
+      Your job is to search for the latest AI news, identify the most important developments,
+      and provide clear, concise summaries with sources.
+      
+      Focus on:
+      - Major AI product launches
+      - Research breakthroughs
+      - Industry trends
+      - Policy and regulation updates
+    tools:
+      - search_tool
+    verbose: true
+
+# Task to execute on each scheduled run
+task: |
+  Search for the latest AI news from the past 24 hours.
+  Provide the top 3 most important stories with:
+  1. Headline
+  2. Brief summary (2-3 sentences)
+  3. Source URL
+  4. Why it's significant
+
+# Schedule configuration
+schedule:
+  # Interval options:
+  # - "hourly" - Every hour
+  # - "daily" - Every 24 hours
+  # - "*/30m" - Every 30 minutes
+  # - "*/6h" - Every 6 hours
+  # - "3600" - Custom seconds
+  interval: "hourly"
+  
+  # Maximum retry attempts if execution fails
+  max_retries: 3
+  
+  # Run immediately when scheduler starts (before first scheduled interval)
+  run_immediately: true
+
+# Optional: Variables for customization
+variables:
+  news_category: "AI and Machine Learning"
+  region: "Global"
+  time_range: "24 hours"
@@ -0,0 +1,69 @@
+"""
+Accuracy Evaluation Example
+
+This example demonstrates how to evaluate agent output accuracy
+by comparing against expected outputs using LLM-as-judge.
+"""
+
+import os
+from praisonaiagents import Agent
+from praisonaiagents.eval import AccuracyEvaluator
+
+# Check if we have an API key
+has_api_key = os.getenv("OPENAI_API_KEY") is not None
+
+if has_api_key:
+    # Create a simple math tutor agent
+    agent = Agent(
+        instructions="You are a math tutor. Answer math questions concisely with just the number."
+    )
+
+    # Create accuracy evaluator
+    evaluator = AccuracyEvaluator(
+        agent=agent,
+        input_text="What is 2 + 2?",
+        expected_output="4",
+        num_iterations=3,  # Run 3 times for statistical significance
+        verbose=True
+    )
+
+    # Run evaluation
+    result = evaluator.run(print_summary=True)
+
+    # Check results
+    print(f"\nAverage Score: {result.avg_score}/10")
+    print(f"Passed: {result.passed}")
+else:
+    print("⚠️  No OPENAI_API_KEY found. Using mock evaluation...")
+    
+    # Create evaluator without agent for mock testing
+    evaluator = AccuracyEvaluator(
+        func=lambda x: "4",  # Mock function
+        input_text="What is 2 + 2?",
+        expected_output="4",
+        num_iterations=1,
+        verbose=True
+    )
+    
+    # Test with mock output
+    try:
+        result = evaluator.run(print_summary=True)
+        print(f"\nMock Average Score: {result.avg_score}/10")
+        print(f"Mock Passed: {result.passed}")
+    except Exception as e:
+        print(f"Mock evaluation failed: {e}")
+        print("This is expected without an API key")
+
+# You can also evaluate pre-generated outputs
+print("\n--- Testing pre-generated output evaluation ---")
+output = "The answer is 4"
+
+if has_api_key:
+    try:
+        result2 = evaluator.evaluate_output(output, print_summary=True)
+        print(f"Pre-generated output score: {result2.avg_score}/10")
+    except Exception as e:
+        print(f"Pre-generated evaluation failed: {e}")
+else:
+    print("⚠️  Skipping pre-generated evaluation (no API key)")
+    print("To run full evaluation, set OPENAI_API_KEY environment variable")
@@ -0,0 +1,105 @@
+"""
+Batch Evaluation Example
+
+This example demonstrates how to run batch evaluations
+from a JSON test file.
+"""
+
+import os
+import json
+import tempfile
+from praisonaiagents.eval import AccuracyEvaluator
+
+# Check if we have an API key
+has_api_key = os.getenv("OPENAI_API_KEY") is not None
+
+# Create a test file with multiple test cases
+test_cases = [
+    {
+        "input": "What is 2 + 2?",
+        "expected": "4"
+    },
+    {
+        "input": "What is the capital of France?",
+        "expected": "Paris"
+    },
+    {
+        "input": "What color is the sky?",
+        "expected": "Blue"
+    }
+]
+
+# Save test cases to a temporary file
+with tempfile.NamedTemporaryFile(mode='w', suffix='.json', delete=False) as f:
+    json.dump(test_cases, f)
+    test_file = f.name
+
+print(f"Test file created: {test_file}")
+print(f"Running {len(test_cases)} test cases...")
+
+if has_api_key:
+    # You can use the CLI for batch evaluation:
+    # praisonai eval batch --agent agents.yaml --test-file tests.json --batch-type accuracy
+
+    # Or programmatically:
+    from praisonaiagents import Agent
+
+    agent = Agent(
+        instructions="You are a helpful assistant. Answer questions concisely."
+    )
+
+    results = []
+    for i, test_case in enumerate(test_cases):
+        print(f"\nTest {i + 1}: {test_case['input']}")
+        
+        evaluator = AccuracyEvaluator(
+            agent=agent,
+            input_text=test_case["input"],
+            expected_output=test_case["expected"],
+            num_iterations=1
+        )
+        
+        result = evaluator.run()
+        results.append({
+            "input": test_case["input"],
+            "expected": test_case["expected"],
+            "score": result.avg_score,
+            "passed": result.passed
+        })
+        
+        print(f"  Score: {result.avg_score}/10")
+        print(f"  Passed: {result.passed}")
+
+    # Summary
+    passed = sum(1 for r in results if r["passed"])
+    print(f"\n{'='*50}")
+    print(f"Summary: {passed}/{len(results)} tests passed")
+    print(f"Pass Rate: {passed/len(results):.1%}")
+else:
+    print("⚠️  No OPENAI_API_KEY found. Running mock batch evaluation...")
+    
+    # Mock evaluation without API key
+    results = []
+    for i, test_case in enumerate(test_cases):
+        print(f"\nTest {i + 1}: {test_case['input']}")
+        
+        # Mock result (all would fail without API key)
+        mock_score = 0.0
+        mock_passed = False
+        
+        results.append({
+            "input": test_case["input"],
+            "expected": test_case["expected"],
+            "score": mock_score,
+            "passed": mock_passed
+        })
+        
+        print(f"  Score: {mock_score}/10 (mock)")
+        print(f"  Passed: {mock_passed} (expected without API key)")
+
+    # Summary
+    passed = sum(1 for r in results if r["passed"])
+    print(f"\n{'='*50}")
+    print(f"Summary: {passed}/{len(results)} tests passed (mock evaluation)")
+    print(f"Pass Rate: {passed/len(results):.1%}")
+    print("\nTo run real batch evaluation, set OPENAI_API_KEY environment variable")
@@ -0,0 +1,82 @@
+"""
+Criteria Evaluation Example
+
+This example demonstrates how to evaluate agent outputs
+against custom criteria using LLM-as-judge.
+"""
+
+import os
+from praisonaiagents import Agent
+from praisonaiagents.eval import CriteriaEvaluator
+
+# Check if we have an API key
+has_api_key = os.getenv("OPENAI_API_KEY") is not None
+
+if has_api_key:
+    print("--- Testing Criteria Evaluation with Agent ---")
+    # Create a customer service agent
+    agent = Agent(
+        instructions="You are a friendly customer service agent. Be helpful and empathetic."
+    )
+
+    # Create criteria evaluator with numeric scoring
+    evaluator = CriteriaEvaluator(
+        criteria="Response is helpful, empathetic, and provides a clear solution",
+        agent=agent,
+        input_text="I'm frustrated because my order hasn't arrived yet.",
+        scoring_type="numeric",  # Score 1-10
+        threshold=7.0,           # Pass if score >= 7
+        num_iterations=2,
+        verbose=True
+    )
+
+    # Run evaluation
+    result = evaluator.run(print_summary=True)
+
+    print("\nNumeric Scoring Results:")
+    print(f"  Average Score: {result.avg_score}/10")
+    print(f"  Pass Rate: {result.pass_rate:.1%}")
+    print(f"  Passed: {result.passed}")
+
+    # Binary scoring example (pass/fail)
+    print("\n--- Testing Binary Scoring ---")
+    binary_evaluator = CriteriaEvaluator(
+        criteria="Response does not contain any offensive language",
+        agent=agent,
+        input_text="Tell me a joke",
+        scoring_type="binary",  # Pass or Fail
+        verbose=True
+    )
+
+    binary_result = binary_evaluator.run(print_summary=True)
+    
+    # With on_fail callback
+    print("\n--- Testing Failure Callback ---")
+    def handle_failure(score):
+        print(f"ALERT: Evaluation failed with score {score.score}")
+        print(f"Reasoning: {score.reasoning}")
+
+    callback_evaluator = CriteriaEvaluator(
+        criteria="Response is professional and helpful",
+        agent=agent,
+        input_text="Help me",
+        on_fail=handle_failure,
+        threshold=8.0
+    )
+
+    callback_evaluator.run()
+else:
+    print("⚠️  No OPENAI_API_KEY found. Skipping agent-based criteria evaluation...")
+
+# Test pre-generated output evaluation (doesn't need agent)
+print("\n--- Testing Pre-generated Output Evaluation ---")
+if has_api_key:
+    output = "I understand your frustration. Let me check on your order right away."
+    try:
+        result2 = evaluator.evaluate_output(output, print_summary=True)
+        print(f"Pre-generated output score: {result2.avg_score}/10")
+    except Exception as e:
+        print(f"Pre-generated evaluation failed: {e}")
+else:
+    print("⚠️  Skipping pre-generated evaluation (no API key)")
+    print("To run full evaluation, set OPENAI_API_KEY environment variable")
@@ -0,0 +1,59 @@
+"""
+Performance Evaluation Example
+
+This example demonstrates how to benchmark agent performance
+by measuring runtime and memory usage across multiple iterations.
+"""
+
+import os
+from praisonaiagents import Agent
+from praisonaiagents.eval import PerformanceEvaluator
+
+# Check if we have an API key
+has_api_key = os.getenv("OPENAI_API_KEY") is not None
+
+if has_api_key:
+    print("--- Testing Agent Performance ---")
+    # Create a simple agent
+    agent = Agent(
+        instructions="You are a helpful assistant. Answer questions briefly."
+    )
+
+    # Create performance evaluator
+    evaluator = PerformanceEvaluator(
+        agent=agent,
+        input_text="What is the capital of France?",
+        num_iterations=10,  # Run 10 benchmark iterations
+        warmup_runs=2,      # 2 warmup runs before measurement
+        track_memory=True,  # Track memory usage
+        verbose=True
+    )
+
+    # Run evaluation
+    result = evaluator.run(print_summary=True)
+
+    # Access detailed metrics
+    print("\nAgent Performance Results:")
+    print(f"  Average Time: {result.avg_run_time:.4f}s")
+    print(f"  Min Time: {result.min_run_time:.4f}s")
+    print(f"  Max Time: {result.max_run_time:.4f}s")
+    print(f"  Median Time: {result.median_run_time:.4f}s")
+    print(f"  P95 Time: {result.p95_run_time:.4f}s")
+    print(f"  Avg Memory: {result.avg_memory:.2f} MB")
+else:
+    print("⚠️  No OPENAI_API_KEY found. Skipping agent performance test...")
+
+# You can also benchmark any function
+def my_function():
+    import time
+    time.sleep(0.1)
+    return "done"
+
+func_evaluator = PerformanceEvaluator(
+    func=my_function,
+    num_iterations=5,
+    warmup_runs=1,
+    track_memory=False
+)
+
+func_result = func_evaluator.run(print_summary=True)
@@ -0,0 +1,72 @@
+"""
+Reliability Evaluation Example
+
+This example demonstrates how to verify that agents call
+the expected tools during execution.
+"""
+
+import os
+from praisonaiagents import Agent
+from praisonaiagents.eval import ReliabilityEvaluator
+
+
+def search_web(query: str) -> str:
+    """Search the web for information."""
+    return f"Search results for: {query}"
+
+
+def calculate(expression: str) -> str:
+    """Calculate a math expression."""
+    return str(eval(expression))
+
+
+# Check if we have an API key
+has_api_key = os.getenv("OPENAI_API_KEY") is not None
+
+if has_api_key:
+    print("--- Testing Agent Tool Call Reliability ---")
+    # Create an agent with tools
+    agent = Agent(
+        instructions="You are a helpful assistant with access to search and calculator tools.",
+        tools=[search_web, calculate]
+    )
+
+    # Create reliability evaluator
+    evaluator = ReliabilityEvaluator(
+        agent=agent,
+        input_text="Search for the weather in Paris and calculate 25 * 4",
+        expected_tools=["search_web", "calculate"],  # Tools that should be called
+        forbidden_tools=["delete_file"],  # Tools that should NOT be called
+        verbose=True
+    )
+
+    # Run evaluation
+    result = evaluator.run(print_summary=True)
+
+    # Check results
+    print("\nAgent Reliability Results:")
+    print(f"  Passed: {result.passed}")
+    print(f"  Pass Rate: {result.pass_rate:.1%}")
+else:
+    print("⚠️  No OPENAI_API_KEY found. Skipping agent reliability test...")
+    print("Agent would fail to call tools without API key (expected behavior)")
+
+# You can also evaluate pre-recorded tool calls (doesn't need API key)
+print("\n--- Testing Pre-recorded Tool Call Evaluation ---")
+# Create a mock agent for tool call evaluation
+mock_agent = Agent(instructions="Mock agent")
+evaluator = ReliabilityEvaluator(
+    agent=mock_agent,
+    expected_tools=["search_web", "calculate"],  # Tools that should be called
+    forbidden_tools=["delete_file"],  # Tools that should NOT be called
+    verbose=True
+)
+
+result2 = evaluator.evaluate_tool_calls(
+    actual_tools=["search_web", "calculate"],
+    print_summary=True
+)
+
+print("\nPre-recorded Tool Call Results:")
+print(f"  Passed: {result2.passed}")
+print(f"  Pass Rate: {result2.pass_rate:.1%}")
@@ -0,0 +1,132 @@
+"""
+AutoAgents Workflow Patterns Example
+
+This example demonstrates the new workflow auto-generation features:
+- Dynamic agent count based on task complexity
+- Multiple workflow patterns (sequential, parallel, routing, orchestrator-workers, evaluator-optimizer)
+- Pattern recommendation based on task keywords
+
+Documentation: https://docs.praison.ai/features/autoagents
+CLI Reference: https://docs.praison.ai/nocode/auto
+"""
+
+from praisonai.auto import AutoGenerator, WorkflowAutoGenerator
+import os
+import tempfile
+
+# Ensure API key is set
+if not os.environ.get("OPENAI_API_KEY"):
+    print("Please set OPENAI_API_KEY environment variable")
+    exit(1)
+
+print("=" * 60)
+print("AutoAgents Workflow Patterns Demo")
+print("=" * 60)
+
+# =============================================================================
+# Example 1: Dynamic Agent Count (Simple vs Complex Tasks)
+# =============================================================================
+print("\n📊 Example 1: Dynamic Agent Count")
+print("-" * 40)
+
+# Simple task - should create 1-2 agents
+with tempfile.NamedTemporaryFile(mode='w', suffix='.yaml', delete=False) as f:
+    simple_file = f.name
+
+simple_generator = AutoGenerator(
+    topic="Write a haiku about spring",
+    agent_file=simple_file,
+    framework="praisonai"
+)
+
+print("Simple task: 'Write a haiku about spring'")
+result = simple_generator.generate()
+print(f"Generated: {result}")
+
+# Complex task - should create 3-4 agents
+with tempfile.NamedTemporaryFile(mode='w', suffix='.yaml', delete=False) as f:
+    complex_file = f.name
+
+complex_generator = AutoGenerator(
+    topic="Research AI trends, analyze market data, write a comprehensive report with executive summary",
+    agent_file=complex_file,
+    framework="praisonai"
+)
+
+print("\nComplex task: 'Research AI trends, analyze market data...'")
+result = complex_generator.generate()
+print(f"Generated: {result}")
+
+# =============================================================================
+# Example 2: Pattern Recommendation
+# =============================================================================
+print("\n🎯 Example 2: Pattern Recommendation")
+print("-" * 40)
+
+test_tasks = [
+    ("Write a blog post", "sequential"),
+    ("Research from multiple sources simultaneously", "parallel"),
+    ("Classify and route customer requests", "routing"),
+    ("Comprehensive analysis and break down the problem", "orchestrator-workers"),
+    ("Refine and improve content quality", "evaluator-optimizer"),
+]
+
+for task, expected in test_tasks:
+    generator = WorkflowAutoGenerator(topic=task)
+    recommended = generator.recommend_pattern()
+    status = "✅" if recommended == expected else "⚠️"
+    print(f"{status} Task: '{task[:40]}...'")
+    print(f"   Recommended: {recommended}")
+
+# =============================================================================
+# Example 3: Generate Workflow with Specific Pattern
+# =============================================================================
+print("\n🔄 Example 3: Generate Workflows with Patterns")
+print("-" * 40)
+
+patterns = ["sequential", "parallel", "orchestrator-workers", "evaluator-optimizer"]
+
+for pattern in patterns:
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.yaml', delete=False) as f:
+        workflow_file = f.name
+    
+    generator = WorkflowAutoGenerator(
+        topic="Create a market analysis report",
+        workflow_file=workflow_file
+    )
+    
+    print(f"\n📋 Pattern: {pattern}")
+    result = generator.generate(pattern=pattern)
+    print(f"   Generated: {result}")
+    
+    # Read and show structure
+    import yaml
+    with open(result, 'r') as f:
+        data = yaml.safe_load(f)
+    print(f"   Agents: {list(data.get('agents', {}).keys())}")
+    print(f"   Steps: {len(data.get('steps', []))}")
+
+# =============================================================================
+# Example 4: CLI Usage (for reference)
+# =============================================================================
+print("\n💻 Example 4: CLI Commands (for reference)")
+print("-" * 40)
+print("""
+# Auto-generate agents (dynamic count)
+praisonai --auto "Write a haiku about spring"
+
+# Auto-generate workflow with pattern
+praisonai workflow auto "Research and write a report" --pattern sequential
+praisonai workflow auto "Research from multiple sources" --pattern parallel
+praisonai workflow auto "Comprehensive market analysis" --pattern orchestrator-workers
+praisonai workflow auto "Refine content quality" --pattern evaluator-optimizer
+
+# With output file
+praisonai workflow auto "Customer support routing" --pattern routing --output support.yaml
+""")
+
+print("\n" + "=" * 60)
+print("✅ AutoAgents Workflow Patterns Demo Complete!")
+print("=" * 60)
+print("\nDocumentation: https://docs.praison.ai/features/autoagents")
+print("CLI Reference: https://docs.praison.ai/nocode/auto")
@@ -0,0 +1,38 @@
+"""
+Autonomy Modes Example for PraisonAI CLI.
+
+Control AI autonomy: suggest, auto_edit, full_auto.
+Docs: https://docs.praison.ai/cli/autonomy-modes
+"""
+
+from praisonai.cli.features import AutonomyModeHandler
+from praisonai.cli.features.autonomy_mode import ActionRequest, ActionType
+
+# Initialize with suggest mode (safest)
+handler = AutonomyModeHandler()
+manager = handler.initialize(mode="suggest")
+
+print(f"Current mode: {handler.get_mode()}")
+print("Available modes: suggest, auto_edit, full_auto")
+
+# Create an action request
+action = ActionRequest(
+    action_type=ActionType.FILE_WRITE,
+    description="Edit src/main.py to add logging",
+    details={"file": "src/main.py"}
+)
+
+# In suggest mode, this would require approval
+# For demo, we'll just show the action
+print(f"\nAction: {action.description}")
+print(f"Type: {action.action_type.value}")
+print("Would require approval in 'suggest' mode")
+
+# Change to auto_edit mode
+handler.set_mode("auto_edit")
+print(f"\nChanged to: {handler.get_mode()}")
+print("File edits now auto-approved, shell commands still require approval")
+
+# Get statistics
+stats = handler.get_stats()
+print(f"\nStats: {stats}")
@@ -0,0 +1,34 @@
+"""
+Cost Tracking Example for PraisonAI CLI.
+
+Real-time token usage and cost monitoring.
+Docs: https://docs.praison.ai/cli/cost-tracking
+"""
+
+from praisonai.cli.features import CostTrackerHandler
+
+# Initialize tracker
+handler = CostTrackerHandler()
+tracker = handler.initialize(session_id="demo-session")
+
+# Track some requests
+handler.track_request("gpt-4o", input_tokens=1000, output_tokens=500)
+handler.track_request("gpt-4o-mini", input_tokens=2000, output_tokens=800)
+handler.track_request("claude-3-5-sonnet", input_tokens=1500, output_tokens=600)
+
+# Get totals
+print(f"Total tokens: {handler.get_tokens():,}")
+print(f"Total cost: ${handler.get_cost():.4f}")
+
+# Get detailed summary
+summary = handler.get_summary()
+print("\n=== Session Summary ===")
+print(f"Session ID: {summary['session_id']}")
+print(f"Total requests: {summary['total_requests']}")
+print(f"Input tokens: {summary['total_input_tokens']:,}")
+print(f"Output tokens: {summary['total_output_tokens']:,}")
+print(f"Total cost: ${summary['total_cost']:.4f}")
+print(f"Avg cost/request: ${summary['avg_cost_per_request']:.4f}")
+
+# Display formatted (uses Rich if available)
+handler.display_summary()
@@ -0,0 +1,45 @@
+"""
+Git Integration Example for PraisonAI CLI.
+
+Auto-commit with AI messages, diff viewing, undo.
+Docs: https://docs.praison.ai/cli/git-integration
+"""
+
+from praisonai.cli.features import GitIntegrationHandler
+
+# Initialize
+handler = GitIntegrationHandler(verbose=True)
+git = handler.initialize(repo_path=".")
+
+# Check if it's a git repo
+if not git.is_repo:
+    print("Not a git repository!")
+    exit(1)
+
+# Show status
+print("=== Git Status ===")
+status = handler.show_status()
+print(f"Branch: {status.branch}")
+print(f"Staged: {len(status.staged_files)}")
+print(f"Modified: {len(status.modified_files)}")
+print(f"Untracked: {len(status.untracked_files)}")
+
+# Show diff (if any changes)
+if status.has_changes:
+    print("\n=== Diff ===")
+    diff = handler.show_diff()
+    if diff:
+        print(diff[:300] + "..." if len(diff) > 300 else diff)
+
+# Show recent commits
+print("\n=== Recent Commits ===")
+commits = handler.show_log(count=5)
+for c in commits:
+    print(f"  {c.short_hash} {c.message[:50]}")
+
+# To commit with AI message:
+# commit = handler.commit()  # Auto-generates message
+# commit = handler.commit(message="Custom message")
+
+# To undo last commit:
+# handler.undo(soft=True)  # Keeps changes staged
@@ -0,0 +1,49 @@
+"""
+Interactive TUI Example for PraisonAI CLI.
+
+Rich terminal interface with completions and history.
+Docs: https://docs.praison.ai/cli/interactive-tui
+"""
+
+from praisonai.cli.features import InteractiveTUIHandler
+from praisonai.cli.features.interactive_tui import InteractiveConfig
+
+# Custom configuration
+config = InteractiveConfig(
+    prompt="ai> ",
+    multiline=True,
+    enable_completions=True,
+    show_status_bar=True
+)
+
+# Define callbacks
+def on_input(text):
+    """Handle regular input."""
+    return f"You said: {text}"
+
+def on_command(cmd):
+    """Handle slash commands."""
+    if cmd == "/exit":
+        return {"type": "exit"}
+    return {"type": "command", "message": f"Executed: {cmd}"}
+
+# Initialize
+handler = InteractiveTUIHandler()
+session = handler.initialize(
+    config=config,
+    on_input=on_input,
+    on_command=on_command
+)
+
+# Add commands for completion
+session.add_commands(["help", "exit", "cost", "model", "plan"])
+
+# Add symbols from codebase
+session.add_symbols(["MyClass", "my_function"])
+
+print("Interactive TUI initialized!")
+print("Commands available: /help, /exit, /cost, /model, /plan")
+print("Run handler.run() to start interactive session")
+
+# Uncomment to run interactive session:
+# handler.run()
@@ -0,0 +1,24 @@
+"""
+Repository Map Example for PraisonAI CLI.
+
+Intelligent codebase mapping with symbol extraction.
+Docs: https://docs.praison.ai/cli/repo-map
+"""
+
+from praisonai.cli.features import RepoMapHandler
+
+# Initialize with current directory
+handler = RepoMapHandler(verbose=True)
+repo_map = handler.initialize(root=".")
+
+# Get the repository map
+print("=== Repository Map ===")
+map_str = handler.get_map()
+print(map_str[:500] if len(map_str) > 500 else map_str)
+print("..." if len(map_str) > 500 else "")
+
+# Refresh after file changes
+handler.refresh()
+
+# Get symbols for a specific file (if exists)
+# symbols = handler.get_context("MyClass")
@@ -0,0 +1,42 @@
+"""
+Sandbox Execution Example for PraisonAI CLI.
+
+Secure isolated command execution (--sandbox flag).
+Docs: https://docs.praison.ai/cli/sandbox-execution
+"""
+
+from praisonai.cli.features import SandboxExecutorHandler
+from praisonai.cli.features.sandbox_executor import SandboxMode
+
+# Initialize with basic sandbox
+handler = SandboxExecutorHandler(verbose=True)
+sandbox = handler.initialize(mode="basic")
+
+print(f"Sandbox mode: {handler.get_mode()}")
+print(f"Sandbox enabled: {handler.is_enabled}")
+
+# Execute a safe command
+print("\n=== Safe Command ===")
+result = handler.execute("echo 'Hello from sandbox!'")
+print(f"Success: {result.success}")
+print(f"Output: {result.stdout.strip()}")
+print(f"Sandboxed: {result.was_sandboxed}")
+
+# Try a blocked command
+print("\n=== Blocked Command ===")
+result = handler.execute("rm -rf /")
+print(f"Success: {result.success}")
+print(f"Violations: {result.policy_violations}")
+
+# Validate command before execution
+print("\n=== Command Validation ===")
+violations = handler.validate_command("sudo apt install something")
+if violations:
+    print(f"Command blocked: {violations}")
+else:
+    print("Command allowed")
+
+# Available modes
+print("\n=== Available Modes ===")
+for mode in SandboxMode:
+    print(f"  {mode.value}")
@@ -0,0 +1,51 @@
+"""
+Slash Commands Example for PraisonAI CLI.
+
+Demonstrates interactive slash commands like /help, /cost, /model.
+Docs: https://docs.praison.ai/cli/slash-commands
+"""
+
+from praisonai.cli.features import SlashCommandHandler
+from praisonai.cli.features.slash_commands import SlashCommand, CommandKind
+
+# Create handler
+handler = SlashCommandHandler()
+
+# Execute commands
+print("=== /help ===")
+result = handler.execute("/help")
+print(result.get("message", ""))
+
+print("\n=== /cost ===")
+result = handler.execute("/cost")
+print(result.get("message", ""))
+
+print("\n=== /model ===")
+result = handler.execute("/model")
+print(result.get("message", ""))
+
+# Check if input is a command
+print("\n=== Command Detection ===")
+print(f"/help is command: {handler.is_command('/help')}")
+print(f"hello is command: {handler.is_command('hello')}")
+
+# Get completions for auto-complete
+print("\n=== Completions for '/he' ===")
+completions = handler.get_completions("/he")
+print(completions)
+
+# Register custom command
+def my_custom_handler(args, context):
+    return {"type": "custom", "message": f"Custom command with args: {args}"}
+
+custom_cmd = SlashCommand(
+    name="mycommand",
+    description="My custom command",
+    handler=my_custom_handler,
+    kind=CommandKind.ACTION,
+    aliases=["mc"]
+)
+
+handler.register(custom_cmd)
+result = handler.execute("/mycommand arg1 arg2")
+print(f"\n=== Custom Command ===\n{result.get('message', '')}")
@@ -0,0 +1,34 @@
+"""
+Claude Code CLI Integration Example
+
+Demonstrates how to use Claude Code CLI as an agent tool in PraisonAI.
+"""
+
+import asyncio
+from praisonai.integrations import ClaudeCodeIntegration
+
+async def main():
+    # Create Claude Code integration
+    claude = ClaudeCodeIntegration(
+        workspace=".",
+        output_format="json",
+        skip_permissions=True,
+        system_prompt="Be concise and follow best practices",
+        allowed_tools=["Read", "Write", "Bash"]
+    )
+    
+    # Check availability
+    print(f"Claude CLI available: {claude.is_available}")
+    print(f"Claude SDK available: {claude.sdk_available}")
+    
+    # Execute a coding task
+    result = await claude.execute("List all Python files in the current directory")
+    print(f"Result: {result}")
+    
+    # Stream output
+    print("\nStreaming output:")
+    async for event in claude.stream("Explain the project structure"):
+        print(event)
+
+if __name__ == "__main__":
+    asyncio.run(main())
@@ -0,0 +1,60 @@
+"""
+Code Editing Tools Example
+
+Demonstrates how to use PraisonAI's code editing tools with SEARCH/REPLACE diffs.
+"""
+
+from praisonai.code import (
+    set_workspace,
+    code_read_file,
+    code_write_file,
+    code_apply_diff,
+    code_list_files,
+    code_execute_command,
+    CODE_TOOLS
+)
+
+# Set workspace
+set_workspace(".")
+
+# List files
+print("Files in current directory:")
+files = code_list_files(".", recursive=False)
+for f in files:
+    print(f"  {f}")
+
+# Read a file
+print("\nReading example file:")
+try:
+    content = code_read_file("README.md", start_line=1, end_line=10)
+    print(content)
+except Exception as e:
+    print(f"  File not found: {e}")
+
+# Write a file
+print("\nWriting test file:")
+code_write_file("test_output.txt", "Hello from PraisonAI Code Tools!")
+print("  Created test_output.txt")
+
+# Apply a diff
+print("\nApplying diff:")
+diff = """
+<<<<<<< SEARCH
+Hello from PraisonAI Code Tools!
+=======
+Hello from PraisonAI Code Tools!
+This line was added by a diff.
+>>>>>>> REPLACE
+"""
+result = code_apply_diff("test_output.txt", diff)
+print(f"  {result}")
+
+# Execute command
+print("\nExecuting command:")
+output = code_execute_command("cat test_output.txt")
+print(f"  {output}")
+
+# Cleanup
+import os
+os.remove("test_output.txt")
+print("\nCleanup complete.")
@@ -0,0 +1,36 @@
+"""
+Codex CLI Integration Example
+
+Demonstrates how to use OpenAI Codex CLI as an agent tool in PraisonAI.
+"""
+
+import asyncio
+from praisonai.integrations import CodexCLIIntegration
+
+async def main():
+    # Create Codex CLI integration
+    codex = CodexCLIIntegration(
+        workspace=".",
+        full_auto=True,
+        sandbox="danger-full-access",
+        json_output=True
+    )
+    
+    # Check availability
+    print(f"Codex CLI available: {codex.is_available}")
+    print(f"Full Auto: {codex.full_auto}")
+    print(f"Sandbox: {codex.sandbox}")
+    
+    # Execute a coding task
+    result = await codex.execute("List files in the current directory")
+    print(f"Result: {result}")
+    
+    # Stream JSON events
+    print("\nStreaming events:")
+    async for event in codex.stream("Explain the project"):
+        event_type = event.get("type", "")
+        if event_type == "item.completed":
+            print(f"  {event}")
+
+if __name__ == "__main__":
+    asyncio.run(main())
@@ -0,0 +1,37 @@
+"""
+Cursor CLI Integration Example
+
+Demonstrates how to use Cursor CLI as an agent tool in PraisonAI.
+"""
+
+import asyncio
+from praisonai.integrations import CursorCLIIntegration
+
+async def main():
+    # Create Cursor CLI integration
+    cursor = CursorCLIIntegration(
+        workspace=".",
+        output_format="json",
+        force=True,
+        model="gpt-5",
+        stream_partial=True
+    )
+    
+    # Check availability
+    print(f"Cursor CLI available: {cursor.is_available}")
+    print(f"Force mode: {cursor.force}")
+    print(f"Model: {cursor.model}")
+    
+    # Execute a coding task
+    result = await cursor.execute("List files in the current directory")
+    print(f"Result: {result}")
+    
+    # Stream output
+    print("\nStreaming output:")
+    async for event in cursor.stream("Explain the project"):
+        content = event.get("content", "")
+        if content:
+            print(content, end="", flush=True)
+
+if __name__ == "__main__":
+    asyncio.run(main())
@@ -0,0 +1,58 @@
+"""
+External Agents Integration Example
+
+Demonstrates how to use all external CLI integrations as agent tools.
+"""
+
+import asyncio
+from praisonai.integrations import (
+    ClaudeCodeIntegration,
+    GeminiCLIIntegration,
+    CodexCLIIntegration,
+    CursorCLIIntegration,
+    get_available_integrations
+)
+
+async def main():
+    # Check availability of all integrations
+    print("=== External CLI Integrations ===\n")
+    
+    availability = get_available_integrations()
+    print("Availability:")
+    for name, avail in availability.items():
+        status = "✅" if avail else "❌"
+        print(f"  {status} {name}")
+    
+    # Create integrations
+    print("\n--- Claude Code ---")
+    claude = ClaudeCodeIntegration(workspace=".", skip_permissions=True)
+    print(f"CLI: {claude.cli_command}")
+    print(f"Available: {claude.is_available}")
+    tool = claude.as_tool()
+    print(f"Tool: {tool.__name__}")
+    
+    print("\n--- Gemini CLI ---")
+    gemini = GeminiCLIIntegration(workspace=".", model="gemini-2.5-flash")
+    print(f"CLI: {gemini.cli_command}")
+    print(f"Available: {gemini.is_available}")
+    tool = gemini.as_tool()
+    print(f"Tool: {tool.__name__}")
+    
+    print("\n--- Codex CLI ---")
+    codex = CodexCLIIntegration(workspace=".", full_auto=True)
+    print(f"CLI: {codex.cli_command}")
+    print(f"Available: {codex.is_available}")
+    tool = codex.as_tool()
+    print(f"Tool: {tool.__name__}")
+    
+    print("\n--- Cursor CLI ---")
+    cursor = CursorCLIIntegration(workspace=".", force=True)
+    print(f"CLI: {cursor.cli_command}")
+    print(f"Available: {cursor.is_available}")
+    tool = cursor.as_tool()
+    print(f"Tool: {tool.__name__}")
+    
+    print("\n=== All integrations ready ===")
+
+if __name__ == "__main__":
+    asyncio.run(main())
@@ -0,0 +1,33 @@
+"""
+Gemini CLI Integration Example
+
+Demonstrates how to use Gemini CLI as an agent tool in PraisonAI.
+"""
+
+import asyncio
+from praisonai.integrations import GeminiCLIIntegration
+
+async def main():
+    # Create Gemini CLI integration
+    gemini = GeminiCLIIntegration(
+        workspace=".",
+        model="gemini-2.5-flash",
+        output_format="json",
+        include_directories=["../lib", "../docs"]
+    )
+    
+    # Check availability
+    print(f"Gemini CLI available: {gemini.is_available}")
+    print(f"Model: {gemini.model}")
+    
+    # Execute a coding task
+    result = await gemini.execute("Analyze the code structure")
+    print(f"Result: {result}")
+    
+    # Execute with stats
+    result, stats = await gemini.execute_with_stats("Explain main.py")
+    print(f"Result: {result}")
+    print(f"Stats: {stats}")
+
+if __name__ == "__main__":
+    asyncio.run(main())
@@ -1,11 +1,48 @@
+"""
+AutoAgents Example
+
+AutoAgents automatically creates and manages AI agents based on high-level instructions.
+It features dynamic agent count based on task complexity.
+
+Documentation: https://docs.praison.ai/features/autoagents
+CLI Reference: https://docs.praison.ai/nocode/auto
+
+New Features:
+- Dynamic agent count (1-4 based on task complexity)
+- Workflow patterns: sequential, parallel, routing, orchestrator-workers, evaluator-optimizer
+- Pattern recommendation based on task keywords
+- Tool preservation from LLM suggestions
+"""
+
 from praisonaiagents import AutoAgents
 from praisonaiagents.tools import duckduckgo

+# Basic usage - AutoAgents analyzes complexity and creates optimal agents
 agents = AutoAgents(
    instructions="Search for information about AI Agents",
    tools=[duckduckgo],
-    process="sequential",
-    verbose=True
+    process="sequential",  # or "hierarchical"
+    verbose=True,
+    max_agents=3  # Maximum number of agents to create
 )

-agents.start()
+result = agents.start()
+print(result)
+
+# =============================================================================
+# CLI Usage Examples (for reference)
+# =============================================================================
+# 
+# # Auto-generate agents (dynamic count based on complexity)
+# praisonai --auto "Write a haiku about spring"
+# praisonai --auto "Research AI trends, analyze data, write report"
+#
+# # Auto-generate workflow with specific pattern
+# praisonai workflow auto "Research and write" --pattern sequential
+# praisonai workflow auto "Research from multiple sources" --pattern parallel
+# praisonai workflow auto "Comprehensive analysis" --pattern orchestrator-workers
+# praisonai workflow auto "Refine content quality" --pattern evaluator-optimizer
+#
+# # With framework selection
+# praisonai --framework crewai --auto "Create a movie script"
+# praisonai --framework autogen --auto "Create a movie script"
@@ -0,0 +1,115 @@
+"""
+24/7 News Checker Agent Example
+
+This example demonstrates how to create a continuously running agent
+that checks for AI news every hour and prints the results.
+
+Usage:
+    python news_checker_agent.py
+    
+The agent will:
+1. Run immediately on start
+2. Check for latest AI news every 1 hour
+3. Print results to console
+4. Continue running until stopped (Ctrl+C)
+"""
+
+import os
+import sys
+from datetime import datetime
+
+# Add parent directory to path for imports
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), '../../..'))
+
+from praisonaiagents import Agent
+from praisonai.agent_scheduler import AgentScheduler
+from tools import search_tool
+
+
+def on_success(result):
+    """Callback function called when agent execution succeeds."""
+    print("\n" + "="*80)
+    print(f"✅ SUCCESS - {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}")
+    print("="*80)
+    print(result)
+    print("="*80 + "\n")
+
+
+def on_failure(error):
+    """Callback function called when agent execution fails."""
+    print("\n" + "="*80)
+    print(f"❌ FAILURE - {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}")
+    print("="*80)
+    print(f"Error: {error}")
+    print("="*80 + "\n")
+
+
+def main():
+    """Main function to set up and run the scheduled news checker agent."""
+    
+    print("\n🤖 Starting 24/7 News Checker Agent")
+    print("="*80)
+    
+    # Create the agent with web search capability
+    agent = Agent(
+        name="NewsChecker",
+        role="AI News Monitor",
+        goal="Monitor and summarize the latest AI and technology news",
+        backstory="""You are an expert AI news analyst who stays up-to-date 
+        with the latest developments in artificial intelligence, machine learning, 
+        and technology. You provide concise, accurate summaries of important news.""",
+        tools=[search_tool],
+        verbose=True
+    )
+    
+    # Define the task
+    task = """Search for the latest AI news from today. 
+    Focus on major developments, breakthroughs, or announcements.
+    Provide a summary of the top 3-5 most important news items.
+    Include the source and key details for each item."""
+    
+    # Create scheduler with callbacks
+    scheduler = AgentScheduler(
+        agent=agent,
+        task=task,
+        on_success=on_success,
+        on_failure=on_failure
+    )
+    
+    print(f"Agent: {agent.name}")
+    print(f"Task: {task}")
+    print(f"Schedule: Every 1 hour")
+    print("="*80)
+    print("\n⏰ Starting scheduler... (Press Ctrl+C to stop)\n")
+    
+    # Start the scheduler
+    # - "hourly" or "*/1h" = every 1 hour
+    # - run_immediately=True = run once before starting the schedule
+    scheduler.start(
+        schedule_expr="hourly",  # Run every hour
+        max_retries=3,           # Retry up to 3 times on failure
+        run_immediately=True     # Run immediately on start
+    )
+    
+    try:
+        # Keep the main thread alive
+        while scheduler.is_running:
+            import time
+            time.sleep(1)
+            
+    except KeyboardInterrupt:
+        print("\n\n🛑 Stopping scheduler...")
+        scheduler.stop()
+        
+        # Print final statistics
+        stats = scheduler.get_stats()
+        print("\n📊 Final Statistics:")
+        print(f"  Total Executions: {stats['total_executions']}")
+        print(f"  Successful: {stats['successful_executions']}")
+        print(f"  Failed: {stats['failed_executions']}")
+        print(f"  Success Rate: {stats['success_rate']:.1f}%")
+        print("\n✅ Agent stopped successfully\n")
+
+
+if __name__ == "__main__":
+    main()
@@ -0,0 +1,143 @@
+"""
+Live News Checker Agent - 24/7 Scheduled Agent
+
+This example demonstrates a real working agent that checks AI news
+every hour using the AgentScheduler with actual API calls.
+
+Requirements:
+    - OPENAI_API_KEY environment variable set
+    - praisonaiagents package installed
+    
+Usage:
+    python news_checker_live.py
+"""
+
+import os
+import sys
+from datetime import datetime
+
+# Add parent directory to path
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), '../../..'))
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), '../../../..'))
+
+from praisonaiagents import Agent
+from praisonai.scheduler import AgentScheduler
+
+# Import search_tool from root tools.py
+try:
+    from tools import search_tool
+except ImportError:
+    # Fallback: create inline search tool
+    from duckduckgo_search import DDGS
+    
+    def search_tool(query: str) -> list:
+        """Search the web using DuckDuckGo."""
+        try:
+            results = []
+            ddgs = DDGS()
+            for result in ddgs.text(keywords=query, max_results=5):
+                results.append({
+                    "title": result.get("title", ""),
+                    "url": result.get("href", ""),
+                    "snippet": result.get("body", ""),
+                })
+            return results
+        except Exception as e:
+            print(f"Search error: {e}")
+            return []
+
+
+def on_success(result):
+    """Callback when agent execution succeeds."""
+    print("\n" + "="*80)
+    print(f"✅ SUCCESS - {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}")
+    print("="*80)
+    print(result)
+    print("="*80 + "\n")
+
+
+def on_failure(error):
+    """Callback when agent execution fails."""
+    print("\n" + "="*80)
+    print(f"❌ FAILURE - {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}")
+    print("="*80)
+    print(f"Error: {error}")
+    print("="*80 + "\n")
+
+
+def main():
+    """Main function to run the scheduled news checker."""
+    
+    # Check for API key
+    if not os.getenv("OPENAI_API_KEY"):
+        print("❌ Error: OPENAI_API_KEY environment variable not set")
+        print("Please set it with: export OPENAI_API_KEY=your_key_here")
+        sys.exit(1)
+    
+    print("\n🤖 Starting 24/7 AI News Checker Agent")
+    print("="*80)
+    
+    # Create the agent with web search capability
+    agent = Agent(
+        name="AI News Monitor",
+        role="Technology News Analyst",
+        goal="Monitor and summarize the latest AI and technology news",
+        backstory="""You are an expert AI news analyst who stays up-to-date 
+        with the latest developments in artificial intelligence, machine learning, 
+        and technology. You provide concise, accurate summaries of important news.""",
+        tools=[search_tool],
+        verbose=True,
+        self_reflect=False  # Disable reflection for faster execution
+    )
+    
+    # Define the task
+    task = """Search for the latest AI news from today. 
+    Focus on major developments, breakthroughs, or announcements.
+    Provide a summary of the top 3 most important news items.
+    Include the source and key details for each item."""
+    
+    # Create scheduler with callbacks
+    scheduler = AgentScheduler(
+        agent=agent,
+        task=task,
+        on_success=on_success,
+        on_failure=on_failure
+    )
+    
+    print(f"Agent: {agent.name}")
+    print(f"Task: {task}")
+    print(f"Schedule: Every 1 hour")
+    print("="*80)
+    print("\n⏰ Starting scheduler... (Press Ctrl+C to stop)\n")
+    
+    # Start the scheduler
+    # For testing: use "*/2m" (every 2 minutes) instead of "hourly"
+    # For production: use "hourly"
+    scheduler.start(
+        schedule_expr="*/2m",  # Every 2 minutes for testing
+        max_retries=3,
+        run_immediately=True  # Run once immediately
+    )
+    
+    try:
+        # Keep the main thread alive
+        while scheduler.is_running:
+            import time
+            time.sleep(1)
+            
+    except KeyboardInterrupt:
+        print("\n\n🛑 Stopping scheduler...")
+        scheduler.stop()
+        
+        # Print final statistics
+        stats = scheduler.get_stats()
+        print("\n📊 Final Statistics:")
+        print(f"  Total Executions: {stats['total_executions']}")
+        print(f"  Successful: {stats['successful_executions']}")
+        print(f"  Failed: {stats['failed_executions']}")
+        print(f"  Success Rate: {stats['success_rate']:.1f}%")
+        print("\n✅ Agent stopped successfully\n")
+
+
+if __name__ == "__main__":
+    main()
@@ -0,0 +1,61 @@
+"""
+Simple test script for 24/7 Agent Scheduler
+
+This script demonstrates the basic usage of AgentScheduler
+for running agents periodically.
+
+Usage:
+    python simple_test.py
+"""
+
+import sys
+import os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), '../../..'))
+
+from unittest.mock import Mock
+from praisonai.scheduler import AgentScheduler
+
+# Create a mock agent for testing
+mock_agent = Mock()
+mock_agent.name = "TestAgent"
+mock_agent.start = Mock(return_value="Test execution successful!")
+
+# Create scheduler
+scheduler = AgentScheduler(
+    agent=mock_agent,
+    task="Test task execution"
+)
+
+print("🤖 Testing AgentScheduler")
+print("=" * 60)
+
+# Test 1: Execute once
+print("\n✅ Test 1: Execute once")
+result = scheduler.execute_once()
+print(f"Result: {result}")
+
+# Test 2: Get stats
+print("\n✅ Test 2: Get statistics")
+stats = scheduler.get_stats()
+print(f"Stats: {stats}")
+
+# Test 3: Start scheduler (will run every 5 seconds)
+print("\n✅ Test 3: Start scheduler (*/5s)")
+scheduler.start("*/5s", run_immediately=True)
+
+import time
+print("Running for 12 seconds...")
+time.sleep(12)
+
+# Stop scheduler
+print("\n✅ Test 4: Stop scheduler")
+scheduler.stop()
+
+# Final stats
+print("\n📊 Final Statistics:")
+final_stats = scheduler.get_stats()
+for key, value in final_stats.items():
+    print(f"  {key}: {value}")
+
+print("\n✅ All tests passed!")
+print("=" * 60)
@@ -0,0 +1,14 @@
+framework: praisonai
+
+agents:
+  - name: "5 Second Timer"
+    role: "Timer Agent"
+    instructions: "Display the current time"
+    verbose: true
+
+task: "Show current time"
+
+schedule:
+  interval: "*/5s"
+  max_retries: 3
+  run_immediately: true
@@ -0,0 +1,14 @@
+framework: praisonai
+
+agents:
+  - name: "Counter Agent"
+    role: "Counter"
+    instructions: "Count from 1 to 5, incrementing by 1 each time. Just output the numbers clearly."
+    verbose: true
+
+task: "Print numbers 1, 2, 3, 4, 5 (one number per line)"
+
+schedule:
+  interval: "*/5s"
+  max_retries: 3
+  run_immediately: true
@@ -0,0 +1,16 @@
+framework: praisonai
+
+agents:
+  - name: "Counter Agent with Limits"
+    role: "Counter"
+    instructions: "Count from 1 to 5, incrementing by 1 each time. Just output the numbers clearly."
+    verbose: true
+
+task: "Print numbers 1, 2, 3, 4, 5 (one number per line)"
+
+schedule:
+  interval: "*/5s"           # Run every 5 seconds
+  max_retries: 3             # Retry up to 3 times on failure
+  run_immediately: true      # Run once immediately on start
+  timeout: 30                # Max 30 seconds per execution
+  max_cost: 0.01             # Stop after spending $0.01 USD
@@ -0,0 +1,15 @@
+framework: praisonai
+
+agents:
+  - name: "Budget Test Agent"
+    role: "Tester"
+    instructions: "Count from 1 to 3"
+    verbose: true
+
+task: "Print numbers 1, 2, 3"
+
+schedule:
+  interval: "*/3s"
+  max_retries: 2
+  run_immediately: true
+  # No max_cost specified - should default to $1.00
@@ -0,0 +1,73 @@
+"""
+Basic Agent Skills Usage Example
+
+This example demonstrates how to use Agent Skills with PraisonAI Agents.
+Skills are modular capabilities that can be loaded and used by agents.
+"""
+
+from praisonaiagents import Agent, SkillManager
+
+# Example 1: Using SkillManager directly
+print("=" * 50)
+print("Example 1: Using SkillManager directly")
+print("=" * 50)
+
+# Create a skill manager
+manager = SkillManager()
+
+# Discover skills from a directory
+# This scans the directory for subdirectories containing SKILL.md files
+skill_count = manager.discover(["./pdf-processing"], include_defaults=False)
+print(f"Discovered {skill_count} skill(s)")
+
+# List available skills
+for skill in manager:
+    print(f"  - {skill.properties.name}: {skill.properties.description[:50]}...")
+
+# Generate prompt XML for system prompt injection
+prompt_xml = manager.to_prompt()
+print("\nGenerated prompt XML:")
+print(prompt_xml)
+
+# Example 2: Using skills with an Agent
+print("\n" + "=" * 50)
+print("Example 2: Using skills with an Agent")
+print("=" * 50)
+
+# Create an agent with skills
+# Skills are lazy-loaded only when accessed
+agent = Agent(
+    name="PDF Assistant",
+    instructions="You are a helpful assistant that can process PDF documents.",
+    skills=["./pdf-processing"],  # Direct skill paths
+)
+
+# Access the skill manager through the agent
+if agent.skill_manager:
+    print(f"Agent has {len(agent.skill_manager)} skill(s) loaded")
+    
+    # Get the skills prompt for system prompt injection
+    skills_prompt = agent.get_skills_prompt()
+    print("\nAgent skills prompt:")
+    print(skills_prompt)
+
+# Example 3: Discovering skills from multiple directories
+print("\n" + "=" * 50)
+print("Example 3: Using skills_dirs for discovery")
+print("=" * 50)
+
+# Create an agent that discovers skills from directories
+agent_with_discovery = Agent(
+    name="Multi-Skill Agent",
+    instructions="You are a versatile assistant with multiple skills.",
+    skills_dirs=["./"],  # Scan current directory for skill subdirectories
+)
+
+if agent_with_discovery.skill_manager:
+    print(f"Discovered {len(agent_with_discovery.skill_manager)} skill(s)")
+    for name in agent_with_discovery.skill_manager.skill_names:
+        print(f"  - {name}")
+
+print("\n" + "=" * 50)
+print("Examples complete!")
+print("=" * 50)
@@ -0,0 +1,163 @@
+"""
+Custom Skill Creation Example
+
+This example demonstrates how to programmatically create and validate
+custom Agent Skills using PraisonAI Agents.
+"""
+
+import tempfile
+from pathlib import Path
+
+from praisonaiagents.skills import (
+    SkillManager,
+    SkillLoader,
+    validate,
+    read_properties,
+)
+
+
+def create_custom_skill(
+    name: str,
+    description: str,
+    instructions: str,
+    base_dir: str = None
+) -> Path:
+    """Create a custom skill programmatically.
+    
+    Args:
+        name: Skill name (kebab-case)
+        description: What the skill does
+        instructions: Markdown instructions for the skill
+        base_dir: Directory to create skill in (default: temp dir)
+        
+    Returns:
+        Path to the created skill directory
+    """
+    if base_dir is None:
+        base_dir = tempfile.mkdtemp()
+    
+    skill_dir = Path(base_dir) / name
+    skill_dir.mkdir(parents=True, exist_ok=True)
+    
+    # Create SKILL.md
+    skill_md_content = f"""---
+name: {name}
+description: {description}
+license: Apache-2.0
+metadata:
+  author: custom
+  version: "1.0"
+---
+
+{instructions}
+"""
+    (skill_dir / "SKILL.md").write_text(skill_md_content)
+    
+    # Create optional directories
+    (skill_dir / "scripts").mkdir(exist_ok=True)
+    (skill_dir / "references").mkdir(exist_ok=True)
+    (skill_dir / "assets").mkdir(exist_ok=True)
+    
+    return skill_dir
+
+
+def main():
+    print("=" * 60)
+    print("Custom Skill Creation Example")
+    print("=" * 60)
+    
+    # Create a custom skill
+    print("\n1. Creating a custom skill...")
+    
+    skill_path = create_custom_skill(
+        name="code-review",
+        description="Review code for best practices, bugs, and improvements. Use when asked to review or analyze code quality.",
+        instructions="""
+# Code Review Skill
+
+## Overview
+This skill enables thorough code review with focus on:
+- Code quality and readability
+- Potential bugs and edge cases
+- Performance considerations
+- Security vulnerabilities
+- Best practices adherence
+
+## Instructions
+
+When reviewing code:
+
+1. **Read the code carefully** - Understand the purpose and flow
+2. **Check for bugs** - Look for null checks, edge cases, error handling
+3. **Evaluate readability** - Variable names, comments, structure
+4. **Consider performance** - Time/space complexity, unnecessary operations
+5. **Security review** - Input validation, injection risks, data exposure
+6. **Suggest improvements** - Provide actionable recommendations
+
+## Output Format
+
+Structure your review as:
+- Summary: Brief overview of the code
+- Issues: List of problems found (severity: high/medium/low)
+- Suggestions: Recommended improvements
+- Positive aspects: What's done well
+"""
+    )
+    
+    print(f"   Created skill at: {skill_path}")
+    
+    # Validate the skill
+    print("\n2. Validating the skill...")
+    errors = validate(skill_path)
+    
+    if errors:
+        print("   Validation errors:")
+        for error in errors:
+            print(f"   - {error}")
+    else:
+        print("   ✓ Skill is valid!")
+    
+    # Load and inspect the skill
+    print("\n3. Loading skill properties...")
+    props = read_properties(skill_path)
+    print(f"   Name: {props.name}")
+    print(f"   Description: {props.description[:60]}...")
+    print(f"   License: {props.license}")
+    
+    # Use with SkillLoader for progressive loading
+    print("\n4. Using SkillLoader for progressive disclosure...")
+    loader = SkillLoader()
+    
+    # Level 1: Load metadata only
+    skill = loader.load_metadata(str(skill_path))
+    print(f"   Level 1 - Metadata loaded: {skill.properties.name}")
+    print(f"   Instructions loaded: {skill.is_activated}")
+    
+    # Level 2: Activate (load instructions)
+    loader.activate(skill)
+    print(f"   Level 2 - Activated: {skill.is_activated}")
+    print(f"   Instructions preview: {skill.instructions[:100]}...")
+    
+    # Use with SkillManager
+    print("\n5. Using with SkillManager...")
+    manager = SkillManager()
+    manager.add_skill(str(skill_path))
+    
+    # Generate prompt XML
+    prompt = manager.to_prompt()
+    print("   Generated prompt XML:")
+    print(prompt)
+    
+    # Clean up
+    print("\n6. Cleanup...")
+    import shutil
+    shutil.rmtree(skill_path.parent)
+    print("   Temporary files removed.")
+    
+    print("\n" + "=" * 60)
+    print("Example complete!")
+    print("=" * 60)
+
+
+if __name__ == "__main__":
+    main()
@@ -0,0 +1,38 @@
+---
+name: pdf-processing
+description: Process and extract information from PDF documents. Use this skill when the user asks to read, analyze, or extract data from PDF files.
+license: Apache-2.0
+compatibility: Works with PraisonAI Agents
+metadata:
+  author: praisonai
+  version: "1.0"
+---
+
+# PDF Processing Skill
+
+## Overview
+
+This skill enables agents to process PDF documents, extract text content, and analyze the information within them.
+
+## When to Use
+
+Activate this skill when:
+- User asks to read or analyze a PDF file
+- User needs to extract text from a PDF
+- User wants to summarize a PDF document
+- User needs to search for information within a PDF
+
+## Instructions
+
+1. First, verify the PDF file exists at the specified path
+2. Use appropriate tools to read the PDF content
+3. Extract text while preserving structure where possible
+4. For large PDFs, process in chunks to manage context
+5. Summarize or analyze based on user's specific request
+
+## Best Practices
+
+- Always confirm the file path before processing
+- Handle encrypted PDFs gracefully with appropriate error messages
+- For scanned PDFs, note that OCR may be required
+- Preserve important formatting like tables and lists
@@ -0,0 +1,90 @@
+"""
+Unified Web Search Example
+
+This example demonstrates the search_web tool that automatically tries
+multiple search providers with fallback support.
+
+Search Provider Priority:
+1. Tavily (requires TAVILY_API_KEY + tavily-python)
+2. Exa (requires EXA_API_KEY + exa_py)
+3. You.com (requires YDC_API_KEY + youdotcom)
+4. DuckDuckGo (requires duckduckgo_search package, no API key)
+5. SearxNG (requires requests + running SearxNG instance)
+
+Installation:
+    pip install praisonaiagents duckduckgo_search
+
+Usage:
+    python web_search_example.py
+"""
+
+from praisonaiagents import Agent
+from praisonaiagents.tools import search_web, get_available_providers
+
+
+def main():
+    print("=" * 60)
+    print("Unified Web Search - Provider Status")
+    print("=" * 60)
+    
+    # Check which providers are available
+    providers = get_available_providers()
+    for p in providers:
+        status = "✓ Available" if p["available"] else f"✗ {p['reason']}"
+        print(f"  {p['name']:12} {status}")
+    
+    print("\n" + "=" * 60)
+    print("Example 1: Basic Search")
+    print("=" * 60)
+    
+    # Simple search - automatically uses best available provider
+    results = search_web("Python programming best practices", max_results=3)
+    
+    if results and "error" not in results[0]:
+        print(f"\nFound {len(results)} results using {results[0].get('provider', 'unknown')}:\n")
+        for r in results:
+            print(f"  Title: {r.get('title', 'N/A')[:60]}...")
+            print(f"  URL: {r.get('url', 'N/A')}")
+            print()
+    else:
+        print(f"\nSearch failed: {results}")
+    
+    print("=" * 60)
+    print("Example 2: Specify Providers")
+    print("=" * 60)
+    
+    # Only try specific providers
+    results = search_web(
+        "machine learning tutorials",
+        max_results=3,
+        providers=["duckduckgo", "tavily"]  # Try these in order
+    )
+    
+    if results and "error" not in results[0]:
+        print(f"\nFound {len(results)} results using {results[0].get('provider', 'unknown')}:\n")
+        for r in results:
+            print(f"  Title: {r.get('title', 'N/A')[:60]}...")
+            print(f"  URL: {r.get('url', 'N/A')}")
+            print()
+    else:
+        print(f"\nSearch failed: {results}")
+    
+    print("=" * 60)
+    print("Example 3: With PraisonAI Agent")
+    print("=" * 60)
+    
+    # Use search_web as an agent tool
+    agent = Agent(
+        name="SearchAgent",
+        role="Web Researcher",
+        goal="Find and summarize information from the web",
+        instructions="Search the web and provide concise summaries of findings.",
+        tools=[search_web]
+    )
+    
+    result = agent.start("What are the top 3 AI trends in 2025?")
+    print(f"\nAgent Response:\n{result}")
+
+
+if __name__ == "__main__":
+    main()
@@ -3,7 +3,7 @@ Praison AI Agents - A package for hierarchical AI agent task execution
 """

 # Apply warning patch BEFORE any imports to intercept warnings at the source
-from . import _warning_patch
+from . import _warning_patch  # noqa: F401

 # Import centralized logging configuration FIRST
 from . import _logging
@@ -78,6 +78,19 @@ def __getattr__(name):
    elif name == "LineRange":
        from praisonaiagents.context.fast import LineRange
        return LineRange
+    # Agent Skills support (lazy loaded for zero performance impact)
+    elif name == "SkillManager":
+        from praisonaiagents.skills import SkillManager
+        return SkillManager
+    elif name == "SkillProperties":
+        from praisonaiagents.skills import SkillProperties
+        return SkillProperties
+    elif name == "SkillMetadata":
+        from praisonaiagents.skills import SkillMetadata
+        return SkillMetadata
+    elif name == "SkillLoader":
+        from praisonaiagents.skills import SkillLoader
+        return SkillLoader
    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")

 # Planning mode support
@@ -289,6 +302,23 @@ __all__ = [
    'ExpandResult'
 ]

+# Add workflow exports
+__all__.extend([
+    'Workflow',
+    'WorkflowStep', 
+    'WorkflowContext',
+    'StepResult',
+    'Route',
+    'Parallel',
+    'Loop',
+    'Repeat',
+    'route',
+    'parallel',
+    'loop',
+    'repeat',
+    'Pipeline'
+])
+
 # Add MCP to __all__ if available
 if _mcp_available:
    __all__.append('MCP')
@@ -305,6 +335,14 @@ __all__.extend([
    'LineRange'
 ])

+# Add Agent Skills exports (lazy loaded)
+__all__.extend([
+    'SkillManager',
+    'SkillProperties',
+    'SkillMetadata',
+    'SkillLoader'
+])
+
 # AG-UI support (optional, lazy loaded)
 try:
    from praisonaiagents.ui.agui import AGUI
@@ -55,10 +55,10 @@ def _get_all_noisy_loggers() -> List[str]:

 def _configure_loggers():
    """Configure all loggers based on LOGLEVEL environment variable."""
-    loglevel = os.environ.get('LOGLEVEL', 'INFO').upper()
+    loglevel = os.environ.get('LOGLEVEL', 'WARNING').upper()
    
-    # When DEBUG is set, allow some HTTP logging for API endpoints
-    if loglevel == 'DEBUG':
+    # When DEBUG or INFO is set, allow some HTTP logging for API endpoints
+    if loglevel in ('DEBUG', 'INFO'):
        allowed_debug_loggers = {"httpx", "httpx._client", "openai._client"}
        
        for logger_name in _get_all_noisy_loggers():
@@ -109,10 +109,10 @@ def configure_root_logger():
    """Configure the root logger with RichHandler."""
    from rich.logging import RichHandler
    
-    loglevel = os.environ.get('LOGLEVEL', 'INFO').upper()
+    loglevel = os.environ.get('LOGLEVEL', 'WARNING').upper()
    
    logging.basicConfig(
-        level=getattr(logging, loglevel, logging.INFO),
+        level=getattr(logging, loglevel, logging.WARNING),
        format="%(asctime)s %(filename)s:%(lineno)d %(levelname)s %(message)s",
        datefmt="[%X]",
        handlers=[RichHandler(rich_tracebacks=True)],
@@ -8,6 +8,7 @@ import functools
 import sys

 # Apply aggressive warning filters first
+warnings.filterwarnings("ignore", message=".*There is no current event loop.*", category=DeprecationWarning)
 warnings.filterwarnings("ignore", message=".*Pydantic serializer warnings.*")
 warnings.filterwarnings("ignore", message=".*PydanticSerializationUnexpectedValue.*")
 warnings.filterwarnings("ignore", message=".*Expected 9 fields but got.*")
@@ -29,7 +30,10 @@ SUPPRESSED_PATTERNS = [
    "PydanticSerializationUnexpectedValue",
    "Expected 9 fields but got 5 for type `Message`",
    "Expected `StreamingChoices` but got `Choices`",
-    "serialized value may not be as expected"
+    "serialized value may not be as expected",
+    "Mixing V1 models and V2 models",
+    "Please upgrade `Settings` to V2",
+    "There is no current event loop"
 ]

@functools.wraps(_original_warn)
@@ -41,7 +45,7 @@ def _patched_warn(message, category=None, stacklevel=1, source=None):
        if pattern in msg_str:
            return
    
-    if category == UserWarning and "pydantic" in msg_str.lower():
+    if category is UserWarning and "pydantic" in msg_str.lower():
        return
    
    _original_warn(message, category, stacklevel, source)
@@ -55,7 +59,7 @@ def _patched_warn_explicit(message, category, filename, lineno, module=None, reg
        if pattern in msg_str:
            return
    
-    if category == UserWarning and "pydantic" in msg_str.lower():
+    if category is UserWarning and "pydantic" in msg_str.lower():
        return
    
    if module and "pydantic" in str(module):
@@ -250,7 +250,9 @@ class Agent:
        fast_context_parallelism: int = 8,
        fast_context_timeout: float = 30.0,
        history_in_context: Optional[int] = None,
-        auto_save: Optional[str] = None
+        auto_save: Optional[str] = None,
+        skills: Optional[List[str]] = None,
+        skills_dirs: Optional[List[str]] = None
    ):
        """Initialize an Agent instance.

@@ -607,6 +609,12 @@ Your Goal: {self.goal}
        self.fast_context_timeout = fast_context_timeout
        self._fast_context_instance = None  # Lazy loaded

+        # Agent Skills configuration (lazy loaded for zero performance impact)
+        self._skills = skills
+        self._skills_dirs = skills_dirs
+        self._skill_manager = None  # Lazy loaded
+        self._skills_initialized = False
+
    @property
    def console(self):
        """Lazily initialize Rich Console only when needed."""
@@ -626,6 +634,35 @@ Your Goal: {self.goal}
    def fast_context_path(self, value):
        self._fast_context_path = value
    
+    @property
+    def skill_manager(self):
+        """Lazily initialize SkillManager only when skills are accessed."""
+        if self._skill_manager is None and (self._skills or self._skills_dirs):
+            from ..skills import SkillManager
+            self._skill_manager = SkillManager()
+            
+            # Add explicit skill paths
+            if self._skills:
+                for skill_path in self._skills:
+                    self._skill_manager.add_skill(skill_path)
+            
+            # Discover skills from directories
+            if self._skills_dirs:
+                self._skill_manager.discover(self._skills_dirs, include_defaults=False)
+            
+            self._skills_initialized = True
+        return self._skill_manager
+    
+    def get_skills_prompt(self) -> str:
+        """Get the XML prompt for available skills.
+        
+        Returns:
+            XML string with <available_skills> block, or empty string if no skills
+        """
+        if self.skill_manager is None:
+            return ""
+        return self.skill_manager.to_prompt()
+    
    @property
    def _openai_client(self):
        """Lazily initialize OpenAI client only when needed."""
@@ -1538,13 +1575,13 @@ Your Goal: {self.goal}"""
        logging.debug(f"{self.name} executing tool {function_name} with arguments: {arguments}")

        # Check if approval is required for this tool
-        from ..approval import is_approval_required, console_approval_callback, get_risk_level, mark_approved, ApprovalDecision
+        from ..approval import is_approval_required, console_approval_callback, get_risk_level, mark_approved, ApprovalDecision, get_approval_callback
        if is_approval_required(function_name):
            risk_level = get_risk_level(function_name)
-            logging.info(f"Tool {function_name} requires approval (risk level: {risk_level})")
+            logging.debug(f"Tool {function_name} requires approval (risk level: {risk_level})")
            
            # Use global approval callback or default console callback
-            callback = approval_callback or console_approval_callback
+            callback = get_approval_callback() or console_approval_callback
            
            try:
                decision = callback(function_name, arguments, risk_level)
@@ -43,6 +43,15 @@ def set_approval_callback(callback_fn: Callable):
    global approval_callback
    approval_callback = callback_fn

+def get_approval_callback() -> Optional[Callable]:
+    """Get the current approval callback function.
+    
+    Returns the custom callback if set, otherwise None.
+    This should be used instead of directly accessing approval_callback
+    to ensure the latest callback is always used.
+    """
+    return approval_callback
+
 def mark_approved(tool_name: str):
    """Mark a tool as approved in the current context."""
    approved = _approved_context.get(set())
@@ -90,7 +99,7 @@ def require_approval(risk_level: RiskLevel = "high"):
            except Exception as e:
                # Fallback to sync approval if async fails
                logging.warning(f"Async approval failed, using sync fallback: {e}")
-                callback = approval_callback or console_approval_callback
+                callback = get_approval_callback() or console_approval_callback
                decision = callback(tool_name, kwargs, risk_level)
            
            if not decision.approved:
@@ -198,7 +207,7 @@ async def request_approval(function_name: str, arguments: Dict[str, Any]) -> App
    risk_level = TOOL_RISK_LEVELS.get(function_name, "medium")
    
    # Use custom callback if set, otherwise use console callback
-    callback = approval_callback or console_approval_callback
+    callback = get_approval_callback() or console_approval_callback
    
    try:
        # Handle async callbacks
@@ -0,0 +1,68 @@
+"""
+PraisonAI Agents Evaluation Framework.
+
+Provides comprehensive evaluation capabilities for AI agents with zero performance
+impact when not in use through lazy loading.
+
+Evaluator Types:
+    - AccuracyEvaluator: Compare output against expected output using LLM-as-judge
+    - PerformanceEvaluator: Measure runtime and memory usage
+    - ReliabilityEvaluator: Verify expected tool calls are made
+    - CriteriaEvaluator: Evaluate against custom criteria
+
+Example:
+    >>> from praisonaiagents.eval import AccuracyEvaluator
+    >>> evaluator = AccuracyEvaluator(
+    ...     agent=my_agent,
+    ...     input_text="What is 2+2?",
+    ...     expected_output="4"
+    ... )
+    >>> result = evaluator.run(print_summary=True)
+"""
+
+__all__ = [
+    "BaseEvaluator",
+    "AccuracyEvaluator",
+    "PerformanceEvaluator",
+    "ReliabilityEvaluator",
+    "CriteriaEvaluator",
+    "AccuracyResult",
+    "PerformanceResult",
+    "ReliabilityResult",
+    "CriteriaResult",
+    "EvaluationScore",
+    "PerformanceMetrics",
+    "ToolCallResult",
+    "CriteriaScore",
+]
+
+_LAZY_IMPORTS = {
+    "BaseEvaluator": ("base", "BaseEvaluator"),
+    "AccuracyEvaluator": ("accuracy", "AccuracyEvaluator"),
+    "PerformanceEvaluator": ("performance", "PerformanceEvaluator"),
+    "ReliabilityEvaluator": ("reliability", "ReliabilityEvaluator"),
+    "CriteriaEvaluator": ("criteria", "CriteriaEvaluator"),
+    "AccuracyResult": ("results", "AccuracyResult"),
+    "PerformanceResult": ("results", "PerformanceResult"),
+    "ReliabilityResult": ("results", "ReliabilityResult"),
+    "CriteriaResult": ("results", "CriteriaResult"),
+    "EvaluationScore": ("results", "EvaluationScore"),
+    "PerformanceMetrics": ("results", "PerformanceMetrics"),
+    "ToolCallResult": ("results", "ToolCallResult"),
+    "CriteriaScore": ("results", "CriteriaScore"),
+}
+
+
+def __getattr__(name: str):
+    """Lazy import mechanism for zero-cost imports when not used."""
+    if name in _LAZY_IMPORTS:
+        module_name, attr_name = _LAZY_IMPORTS[name]
+        import importlib
+        module = importlib.import_module(f".{module_name}", __name__)
+        return getattr(module, attr_name)
+    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
+
+
+def __dir__():
+    """Return list of available attributes for tab completion."""
+    return __all__
@@ -0,0 +1,239 @@
+"""
+Accuracy evaluator for PraisonAI Agents.
+
+Evaluates agent output accuracy by comparing against expected output using LLM-as-judge.
+"""
+
+import os
+import logging
+from typing import Any, Callable, List, Optional, Union, TYPE_CHECKING
+
+from .base import BaseEvaluator
+from .results import AccuracyResult, EvaluationScore
+
+if TYPE_CHECKING:
+    from ..agent.agent import Agent
+    from ..agents.agents import PraisonAIAgents
+
+logger = logging.getLogger(__name__)
+
+
+class AccuracyEvaluator(BaseEvaluator):
+    """
+    Evaluates the accuracy of agent outputs against expected outputs.
+    
+    Uses an LLM as a judge to score outputs on a scale of 1-10.
+    """
+    
+    def __init__(
+        self,
+        agent: Optional[Union["Agent", "PraisonAIAgents"]] = None,
+        func: Optional[Callable[..., str]] = None,
+        input_text: str = "",
+        expected_output: str = "",
+        num_iterations: int = 1,
+        model: Optional[str] = None,
+        name: Optional[str] = None,
+        save_results_path: Optional[str] = None,
+        verbose: bool = False,
+    ):
+        """
+        Initialize the accuracy evaluator.
+        
+        Args:
+            agent: Agent or PraisonAIAgents instance to evaluate
+            func: Alternative callable that returns output string
+            input_text: Input to provide to the agent/function
+            expected_output: Expected output to compare against
+            num_iterations: Number of evaluation iterations
+            model: LLM model to use for judging (defaults to gpt-4o-mini)
+            name: Name for this evaluation
+            save_results_path: Path to save results
+            verbose: Enable verbose output
+        """
+        super().__init__(name=name, save_results_path=save_results_path, verbose=verbose)
+        
+        self.agent = agent
+        self.func = func
+        self.input_text = input_text
+        self.expected_output = expected_output
+        self.num_iterations = num_iterations
+        self.model = model or os.getenv("OPENAI_MODEL_NAME", "gpt-4o-mini")
+        
+        if agent is None and func is None:
+            raise ValueError("Either 'agent' or 'func' must be provided")
+    
+    def _get_output(self) -> str:
+        """Get output from agent or function."""
+        if self.func:
+            return str(self.func(self.input_text))
+        
+        if hasattr(self.agent, 'chat'):
+            return str(self.agent.chat(self.input_text))
+        elif hasattr(self.agent, 'start'):
+            result = self.agent.start(self.input_text)
+            if hasattr(result, 'raw'):
+                return str(result.raw)
+            return str(result)
+        else:
+            raise ValueError("Agent must have 'chat' or 'start' method")
+    
+    def _judge_output(self, output: str) -> EvaluationScore:
+        """
+        Use LLM to judge the output against expected output.
+        
+        Args:
+            output: The actual output to evaluate
+            
+        Returns:
+            EvaluationScore with score and reasoning
+        """
+        try:
+            from openai import OpenAI
+        except ImportError:
+            raise ImportError("OpenAI package required for accuracy evaluation. Install with: pip install openai")
+        
+        client = OpenAI()
+        
+        prompt = f"""You are an expert evaluator. Compare the actual output against the expected output and provide a score from 1-10.
+
+Input: {self.input_text}
+
+Expected Output: {self.expected_output}
+
+Actual Output: {output}
+
+Scoring Guidelines:
+- 10: Perfect match in meaning and completeness
+- 8-9: Very close, minor differences that don't affect correctness
+- 6-7: Mostly correct but missing some details or has minor errors
+- 4-5: Partially correct but significant issues
+- 2-3: Mostly incorrect but shows some understanding
+- 1: Completely wrong or irrelevant
+
+Respond in this exact format:
+SCORE: [number 1-10]
+REASONING: [brief explanation]"""
+
+        response = client.chat.completions.create(
+            model=self.model,
+            messages=[{"role": "user", "content": prompt}],
+            temperature=0.1,
+            max_tokens=200
+        )
+        
+        response_text = response.choices[0].message.content or ""
+        
+        score = 5.0
+        reasoning = "Unable to parse response"
+        
+        lines = response_text.strip().split('\n')
+        for line in lines:
+            if line.startswith('SCORE:'):
+                try:
+                    score = float(line.replace('SCORE:', '').strip())
+                    score = max(1.0, min(10.0, score))
+                except ValueError:
+                    pass
+            elif line.startswith('REASONING:'):
+                reasoning = line.replace('REASONING:', '').strip()
+        
+        return EvaluationScore(
+            score=score,
+            reasoning=reasoning,
+            input_text=self.input_text,
+            output_text=output,
+            expected_output=self.expected_output
+        )
+    
+    def run(self, print_summary: bool = False) -> AccuracyResult:
+        """
+        Execute the accuracy evaluation.
+        
+        Args:
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            AccuracyResult with all evaluation scores
+        """
+        self.before_run()
+        
+        result = AccuracyResult(
+            eval_id=self.eval_id,
+            name=self.name
+        )
+        
+        for i in range(self.num_iterations):
+            if self.verbose:
+                logger.info(f"Running iteration {i + 1}/{self.num_iterations}")
+            
+            try:
+                output = self._get_output()
+                score = self._judge_output(output)
+                result.evaluations.append(score)
+                
+                if self.verbose:
+                    logger.info(f"  Score: {score.score}/10 - {score.reasoning}")
+            except Exception as e:
+                logger.error(f"Error in iteration {i + 1}: {e}")
+                result.evaluations.append(EvaluationScore(
+                    score=0.0,
+                    reasoning=f"Error: {str(e)}",
+                    input_text=self.input_text,
+                    output_text="",
+                    expected_output=self.expected_output
+                ))
+        
+        self.after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
+    
+    async def run_async(self, print_summary: bool = False) -> AccuracyResult:
+        """
+        Execute the accuracy evaluation asynchronously.
+        
+        Args:
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            AccuracyResult with all evaluation scores
+        """
+        await self.async_before_run()
+        result = self.run(print_summary=False)
+        await self.async_after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
+    
+    def evaluate_output(self, output: str, print_summary: bool = False) -> AccuracyResult:
+        """
+        Evaluate a pre-generated output without running the agent.
+        
+        Args:
+            output: The output to evaluate
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            AccuracyResult with evaluation score
+        """
+        self.before_run()
+        
+        result = AccuracyResult(
+            eval_id=self.eval_id,
+            name=self.name
+        )
+        
+        score = self._judge_output(output)
+        result.evaluations.append(score)
+        
+        self.after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
@@ -0,0 +1,143 @@
+"""
+Base evaluator class for PraisonAI Agents evaluation framework.
+
+This module provides the abstract base class that all evaluators inherit from.
+"""
+
+from abc import ABC, abstractmethod
+from typing import Any, Dict, Optional, Union, TYPE_CHECKING
+import uuid
+import logging
+
+if TYPE_CHECKING:
+    from .results import AccuracyResult, PerformanceResult, ReliabilityResult, CriteriaResult
+
+logger = logging.getLogger(__name__)
+
+
+class BaseEvaluator(ABC):
+    """
+    Abstract base class for all evaluators in PraisonAI Agents.
+    
+    Provides common functionality for evaluation lifecycle management,
+    result storage, and telemetry integration.
+    """
+    
+    def __init__(
+        self,
+        name: Optional[str] = None,
+        save_results_path: Optional[str] = None,
+        verbose: bool = False,
+    ):
+        """
+        Initialize the base evaluator.
+        
+        Args:
+            name: Optional name for this evaluation run
+            save_results_path: Optional file path to save results (supports {name}, {eval_id} placeholders)
+            verbose: Enable verbose output during evaluation
+        """
+        self.eval_id = str(uuid.uuid4())[:8]
+        self.name = name or f"eval_{self.eval_id}"
+        self.save_results_path = save_results_path
+        self.verbose = verbose
+        self._result = None
+    
+    def before_run(self) -> None:
+        """
+        Hook called before evaluation runs.
+        Override in subclasses for custom pre-evaluation logic.
+        """
+        if self.verbose:
+            logger.info(f"Starting evaluation: {self.name} (ID: {self.eval_id})")
+    
+    def after_run(self, result: Any) -> None:
+        """
+        Hook called after evaluation completes.
+        Override in subclasses for custom post-evaluation logic.
+        
+        Args:
+            result: The evaluation result
+        """
+        if self.verbose:
+            logger.info(f"Completed evaluation: {self.name}")
+        
+        if self.save_results_path:
+            self._save_result(result)
+    
+    async def async_before_run(self) -> None:
+        """
+        Async hook called before evaluation runs.
+        Override in subclasses for custom async pre-evaluation logic.
+        """
+        self.before_run()
+    
+    async def async_after_run(self, result: Any) -> None:
+        """
+        Async hook called after evaluation completes.
+        Override in subclasses for custom async post-evaluation logic.
+        
+        Args:
+            result: The evaluation result
+        """
+        self.after_run(result)
+    
+    @abstractmethod
+    def run(self, **kwargs) -> Any:
+        """
+        Execute the evaluation synchronously.
+        
+        Returns:
+            Evaluation result (type depends on evaluator)
+        """
+        pass
+    
+    async def run_async(self, **kwargs) -> Any:
+        """
+        Execute the evaluation asynchronously.
+        Default implementation calls sync run().
+        Override for true async evaluation.
+        
+        Returns:
+            Evaluation result (type depends on evaluator)
+        """
+        return self.run(**kwargs)
+    
+    def _save_result(self, result: Any) -> None:
+        """
+        Save evaluation result to file.
+        
+        Args:
+            result: The evaluation result to save
+        """
+        if not self.save_results_path:
+            return
+        
+        try:
+            import json
+            from pathlib import Path
+            
+            file_path = self.save_results_path.format(
+                name=self.name,
+                eval_id=self.eval_id
+            )
+            
+            path = Path(file_path)
+            path.parent.mkdir(parents=True, exist_ok=True)
+            
+            if hasattr(result, 'to_dict'):
+                data = result.to_dict()
+            elif hasattr(result, '__dict__'):
+                data = result.__dict__
+            else:
+                data = {"result": str(result)}
+            
+            path.write_text(json.dumps(data, indent=2, default=str))
+            
+            if self.verbose:
+                logger.info(f"Saved results to: {file_path}")
+        except Exception as e:
+            logger.warning(f"Failed to save results: {e}")
+    
+    def __repr__(self) -> str:
+        return f"{self.__class__.__name__}(name={self.name!r}, eval_id={self.eval_id!r})"
@@ -0,0 +1,289 @@
+"""
+Criteria evaluator for PraisonAI Agents.
+
+Evaluates agent outputs against custom criteria using LLM-as-judge.
+"""
+
+import os
+import logging
+from typing import Callable, List, Literal, Optional, Union, TYPE_CHECKING
+
+from .base import BaseEvaluator
+from .results import CriteriaResult, CriteriaScore
+
+if TYPE_CHECKING:
+    from ..agent.agent import Agent
+    from ..agents.agents import PraisonAIAgents
+
+logger = logging.getLogger(__name__)
+
+
+class CriteriaEvaluator(BaseEvaluator):
+    """
+    Evaluates agent outputs against custom criteria.
+    
+    Uses an LLM as a judge to score outputs based on user-defined criteria.
+    Supports both numeric (1-10) and binary (pass/fail) scoring.
+    """
+    
+    def __init__(
+        self,
+        criteria: str,
+        agent: Optional[Union["Agent", "PraisonAIAgents"]] = None,
+        func: Optional[Callable[..., str]] = None,
+        input_text: str = "",
+        scoring_type: Literal["numeric", "binary"] = "numeric",
+        threshold: float = 7.0,
+        num_iterations: int = 1,
+        model: Optional[str] = None,
+        on_fail: Optional[Callable[[CriteriaScore], None]] = None,
+        name: Optional[str] = None,
+        save_results_path: Optional[str] = None,
+        verbose: bool = False,
+    ):
+        """
+        Initialize the criteria evaluator.
+        
+        Args:
+            criteria: The criteria to evaluate against (e.g., "Response is helpful and accurate")
+            agent: Agent or PraisonAIAgents instance to evaluate
+            func: Alternative callable that returns output string
+            input_text: Input to provide to the agent/function
+            scoring_type: "numeric" (1-10 scale) or "binary" (pass/fail)
+            threshold: Score threshold for passing (numeric mode only)
+            num_iterations: Number of evaluation iterations
+            model: LLM model to use for judging (defaults to gpt-4o-mini)
+            on_fail: Callback function called when evaluation fails
+            name: Name for this evaluation
+            save_results_path: Path to save results
+            verbose: Enable verbose output
+        """
+        super().__init__(name=name, save_results_path=save_results_path, verbose=verbose)
+        
+        self.criteria = criteria
+        self.agent = agent
+        self.func = func
+        self.input_text = input_text
+        self.scoring_type = scoring_type
+        self.threshold = threshold
+        self.num_iterations = num_iterations
+        self.model = model or os.getenv("OPENAI_MODEL_NAME", "gpt-4o-mini")
+        self.on_fail = on_fail
+    
+    def _get_output(self) -> str:
+        """Get output from agent or function."""
+        if self.func:
+            return str(self.func(self.input_text))
+        
+        if self.agent is None:
+            raise ValueError("Either 'agent' or 'func' must be provided")
+        
+        if hasattr(self.agent, 'chat'):
+            return str(self.agent.chat(self.input_text))
+        elif hasattr(self.agent, 'start'):
+            result = self.agent.start(self.input_text)
+            if hasattr(result, 'raw'):
+                return str(result.raw)
+            return str(result)
+        else:
+            raise ValueError("Agent must have 'chat' or 'start' method")
+    
+    def _judge_output(self, output: str) -> CriteriaScore:
+        """
+        Use LLM to judge the output against criteria.
+        
+        Args:
+            output: The output to evaluate
+            
+        Returns:
+            CriteriaScore with score, passed status, and reasoning
+        """
+        try:
+            from openai import OpenAI
+        except ImportError:
+            raise ImportError("OpenAI package required for criteria evaluation. Install with: pip install openai")
+        
+        client = OpenAI()
+        
+        if self.scoring_type == "binary":
+            prompt = f"""You are an expert evaluator. Evaluate the following output against the given criteria.
+
+Criteria: {self.criteria}
+
+Output to evaluate:
+{output}
+
+Determine if the output PASSES or FAILS the criteria.
+
+Respond in this exact format:
+RESULT: [PASS or FAIL]
+REASONING: [brief explanation]"""
+        else:
+            prompt = f"""You are an expert evaluator. Evaluate the following output against the given criteria.
+
+Criteria: {self.criteria}
+
+Output to evaluate:
+{output}
+
+Score the output from 1-10 based on how well it meets the criteria.
+- 10: Perfectly meets all criteria
+- 8-9: Meets criteria very well with minor issues
+- 6-7: Meets most criteria but has some gaps
+- 4-5: Partially meets criteria
+- 2-3: Barely meets criteria
+- 1: Does not meet criteria at all
+
+Respond in this exact format:
+SCORE: [number 1-10]
+REASONING: [brief explanation]"""
+
+        response = client.chat.completions.create(
+            model=self.model,
+            messages=[{"role": "user", "content": prompt}],
+            temperature=0.1,
+            max_tokens=200
+        )
+        
+        response_text = response.choices[0].message.content or ""
+        
+        score = 5.0
+        passed = False
+        reasoning = "Unable to parse response"
+        
+        lines = response_text.strip().split('\n')
+        for line in lines:
+            if self.scoring_type == "binary":
+                if line.startswith('RESULT:'):
+                    result_str = line.replace('RESULT:', '').strip().upper()
+                    passed = result_str == 'PASS'
+                    score = 10.0 if passed else 0.0
+            else:
+                if line.startswith('SCORE:'):
+                    try:
+                        score = float(line.replace('SCORE:', '').strip())
+                        score = max(1.0, min(10.0, score))
+                        passed = score >= self.threshold
+                    except ValueError:
+                        pass
+            
+            if line.startswith('REASONING:'):
+                reasoning = line.replace('REASONING:', '').strip()
+        
+        return CriteriaScore(
+            score=score,
+            passed=passed,
+            reasoning=reasoning,
+            output_text=output,
+            criteria=self.criteria
+        )
+    
+    def run(self, print_summary: bool = False) -> CriteriaResult:
+        """
+        Execute the criteria evaluation.
+        
+        Args:
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            CriteriaResult with all evaluation scores
+        """
+        self.before_run()
+        
+        result = CriteriaResult(
+            criteria=self.criteria,
+            scoring_type=self.scoring_type,
+            threshold=self.threshold,
+            eval_id=self.eval_id,
+            name=self.name
+        )
+        
+        for i in range(self.num_iterations):
+            if self.verbose:
+                logger.info(f"Running iteration {i + 1}/{self.num_iterations}")
+            
+            try:
+                output = self._get_output()
+                score = self._judge_output(output)
+                result.evaluations.append(score)
+                
+                if self.verbose:
+                    status = "PASS" if score.passed else "FAIL"
+                    logger.info(f"  {status} (Score: {score.score}) - {score.reasoning}")
+                
+                if not score.passed and self.on_fail:
+                    self.on_fail(score)
+                    
+            except Exception as e:
+                logger.error(f"Error in iteration {i + 1}: {e}")
+                error_score = CriteriaScore(
+                    score=0.0,
+                    passed=False,
+                    reasoning=f"Error: {str(e)}",
+                    output_text="",
+                    criteria=self.criteria
+                )
+                result.evaluations.append(error_score)
+                
+                if self.on_fail:
+                    self.on_fail(error_score)
+        
+        self.after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
+    
+    async def run_async(self, print_summary: bool = False) -> CriteriaResult:
+        """
+        Execute the criteria evaluation asynchronously.
+        
+        Args:
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            CriteriaResult with all evaluation scores
+        """
+        await self.async_before_run()
+        result = self.run(print_summary=False)
+        await self.async_after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
+    
+    def evaluate_output(self, output: str, print_summary: bool = False) -> CriteriaResult:
+        """
+        Evaluate a pre-generated output without running the agent.
+        
+        Args:
+            output: The output to evaluate
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            CriteriaResult with evaluation score
+        """
+        self.before_run()
+        
+        result = CriteriaResult(
+            criteria=self.criteria,
+            scoring_type=self.scoring_type,
+            threshold=self.threshold,
+            eval_id=self.eval_id,
+            name=self.name
+        )
+        
+        score = self._judge_output(output)
+        result.evaluations.append(score)
+        
+        if not score.passed and self.on_fail:
+            self.on_fail(score)
+        
+        self.after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
@@ -0,0 +1,184 @@
+"""
+Performance evaluator for PraisonAI Agents.
+
+Measures runtime and memory usage of agent executions or functions.
+"""
+
+import time
+import logging
+from typing import Callable, Optional, TYPE_CHECKING
+
+from .base import BaseEvaluator
+from .results import PerformanceResult, PerformanceMetrics
+
+if TYPE_CHECKING:
+    from ..agent.agent import Agent
+    from ..agents.agents import PraisonAIAgents
+
+logger = logging.getLogger(__name__)
+
+
+class PerformanceEvaluator(BaseEvaluator):
+    """
+    Evaluates the performance of agent executions or functions.
+    
+    Measures runtime and memory usage across multiple iterations.
+    """
+    
+    def __init__(
+        self,
+        func: Optional[Callable[[], None]] = None,
+        agent: Optional["Agent"] = None,
+        input_text: str = "",
+        num_iterations: int = 10,
+        warmup_runs: int = 2,
+        track_memory: bool = True,
+        name: Optional[str] = None,
+        save_results_path: Optional[str] = None,
+        verbose: bool = False,
+    ):
+        """
+        Initialize the performance evaluator.
+        
+        Args:
+            func: Callable to benchmark (takes no arguments)
+            agent: Agent to benchmark (will call chat with input_text)
+            input_text: Input text for agent benchmarking
+            num_iterations: Number of benchmark iterations
+            warmup_runs: Number of warmup runs before measurement
+            track_memory: Whether to track memory usage
+            name: Name for this evaluation
+            save_results_path: Path to save results
+            verbose: Enable verbose output
+        """
+        super().__init__(name=name, save_results_path=save_results_path, verbose=verbose)
+        
+        self.func = func
+        self.agent = agent
+        self.input_text = input_text
+        self.num_iterations = num_iterations
+        self.warmup_runs = warmup_runs
+        self.track_memory = track_memory
+        
+        if func is None and agent is None:
+            raise ValueError("Either 'func' or 'agent' must be provided")
+    
+    def _get_memory_usage(self) -> float:
+        """Get current memory usage in MB."""
+        if not self.track_memory:
+            return 0.0
+        
+        try:
+            import tracemalloc
+            current, _ = tracemalloc.get_traced_memory()
+            return current / (1024 * 1024)
+        except Exception:
+            return 0.0
+    
+    def _execute(self) -> None:
+        """Execute the function or agent."""
+        if self.func:
+            self.func()
+        elif self.agent:
+            if hasattr(self.agent, 'chat'):
+                self.agent.chat(self.input_text)
+            elif hasattr(self.agent, 'start'):
+                self.agent.start(self.input_text)
+            else:
+                raise ValueError("Agent must have 'chat' or 'start' method")
+    
+    def run(self, print_summary: bool = False) -> PerformanceResult:
+        """
+        Execute the performance evaluation.
+        
+        Args:
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            PerformanceResult with all performance metrics
+        """
+        self.before_run()
+        
+        result = PerformanceResult(
+            warmup_runs=self.warmup_runs,
+            eval_id=self.eval_id,
+            name=self.name
+        )
+        
+        if self.track_memory:
+            try:
+                import tracemalloc
+                tracemalloc.start()
+            except Exception as e:
+                logger.warning(f"Could not start memory tracking: {e}")
+                self.track_memory = False
+        
+        if self.verbose:
+            logger.info(f"Running {self.warmup_runs} warmup iterations...")
+        
+        for i in range(self.warmup_runs):
+            try:
+                self._execute()
+            except Exception as e:
+                logger.warning(f"Warmup iteration {i + 1} failed: {e}")
+        
+        if self.verbose:
+            logger.info(f"Running {self.num_iterations} benchmark iterations...")
+        
+        for i in range(self.num_iterations):
+            memory_before = self._get_memory_usage()
+            
+            start_time = time.perf_counter()
+            try:
+                self._execute()
+                success = True
+            except Exception as e:
+                logger.error(f"Iteration {i + 1} failed: {e}")
+                success = False
+            end_time = time.perf_counter()
+            
+            memory_after = self._get_memory_usage()
+            
+            if success:
+                metrics = PerformanceMetrics(
+                    run_time=end_time - start_time,
+                    memory_usage=max(0, memory_after - memory_before),
+                    iteration=i + 1
+                )
+                result.metrics.append(metrics)
+                
+                if self.verbose:
+                    logger.info(f"  Iteration {i + 1}: {metrics.run_time:.4f}s, {metrics.memory_usage:.2f}MB")
+        
+        if self.track_memory:
+            try:
+                import tracemalloc
+                tracemalloc.stop()
+            except Exception:
+                pass
+        
+        self.after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
+    
+    async def run_async(self, print_summary: bool = False) -> PerformanceResult:
+        """
+        Execute the performance evaluation asynchronously.
+        
+        Args:
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            PerformanceResult with all performance metrics
+        """
+        await self.async_before_run()
+        result = self.run(print_summary=False)
+        await self.async_after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
@@ -0,0 +1,225 @@
+"""
+Reliability evaluator for PraisonAI Agents.
+
+Evaluates agent reliability by verifying expected tool calls are made.
+"""
+
+import logging
+from typing import Dict, List, Optional, Set, Any, TYPE_CHECKING
+
+from .base import BaseEvaluator
+from .results import ReliabilityResult, ToolCallResult
+
+if TYPE_CHECKING:
+    from ..agent.agent import Agent
+    from ..agents.agents import PraisonAIAgents
+
+logger = logging.getLogger(__name__)
+
+
+class ReliabilityEvaluator(BaseEvaluator):
+    """
+    Evaluates the reliability of agent tool usage.
+    
+    Verifies that expected tools are called during agent execution.
+    """
+    
+    def __init__(
+        self,
+        agent: Optional["Agent"] = None,
+        input_text: str = "",
+        expected_tools: Optional[List[str]] = None,
+        forbidden_tools: Optional[List[str]] = None,
+        name: Optional[str] = None,
+        save_results_path: Optional[str] = None,
+        verbose: bool = False,
+    ):
+        """
+        Initialize the reliability evaluator.
+        
+        Args:
+            agent: Agent to evaluate
+            input_text: Input to provide to the agent
+            expected_tools: List of tool names that should be called
+            forbidden_tools: List of tool names that should NOT be called
+            name: Name for this evaluation
+            save_results_path: Path to save results
+            verbose: Enable verbose output
+        """
+        super().__init__(name=name, save_results_path=save_results_path, verbose=verbose)
+        
+        self.agent = agent
+        self.input_text = input_text
+        self.expected_tools = set(expected_tools or [])
+        self.forbidden_tools = set(forbidden_tools or [])
+        
+        if agent is None:
+            raise ValueError("'agent' must be provided")
+    
+    def _extract_tool_calls(self, response: Any) -> Set[str]:
+        """
+        Extract tool names from agent response.
+        
+        Args:
+            response: Agent response object
+            
+        Returns:
+            Set of tool names that were called
+        """
+        tool_calls = set()
+        
+        if hasattr(response, 'tool_calls') and response.tool_calls:
+            for tc in response.tool_calls:
+                if hasattr(tc, 'name'):
+                    tool_calls.add(tc.name)
+                elif hasattr(tc, 'function') and hasattr(tc.function, 'name'):
+                    tool_calls.add(tc.function.name)
+        
+        if hasattr(self.agent, 'chat_history'):
+            for msg in self.agent.chat_history:
+                if isinstance(msg, dict):
+                    if msg.get('role') == 'assistant' and 'tool_calls' in msg:
+                        for tc in msg.get('tool_calls', []):
+                            if isinstance(tc, dict) and 'function' in tc:
+                                tool_calls.add(tc['function'].get('name', ''))
+                    elif msg.get('role') == 'tool':
+                        tool_name = msg.get('name', '')
+                        if tool_name:
+                            tool_calls.add(tool_name)
+        
+        return tool_calls
+    
+    def run(self, print_summary: bool = False) -> ReliabilityResult:
+        """
+        Execute the reliability evaluation.
+        
+        Args:
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            ReliabilityResult with tool call verification results
+        """
+        self.before_run()
+        
+        result = ReliabilityResult(
+            eval_id=self.eval_id,
+            name=self.name
+        )
+        
+        try:
+            if hasattr(self.agent, 'chat'):
+                response = self.agent.chat(self.input_text)
+            elif hasattr(self.agent, 'start'):
+                response = self.agent.start(self.input_text)
+            else:
+                raise ValueError("Agent must have 'chat' or 'start' method")
+            
+            actual_tools = self._extract_tool_calls(response)
+            
+            if self.verbose:
+                logger.info(f"Expected tools: {self.expected_tools}")
+                logger.info(f"Actual tools called: {actual_tools}")
+            
+            for tool in self.expected_tools:
+                was_called = tool in actual_tools
+                result.tool_results.append(ToolCallResult(
+                    tool_name=tool,
+                    expected=True,
+                    actual=was_called,
+                    passed=was_called
+                ))
+            
+            for tool in self.forbidden_tools:
+                was_called = tool in actual_tools
+                result.tool_results.append(ToolCallResult(
+                    tool_name=tool,
+                    expected=False,
+                    actual=was_called,
+                    passed=not was_called
+                ))
+            
+        except Exception as e:
+            logger.error(f"Error during reliability evaluation: {e}")
+            for tool in self.expected_tools:
+                result.tool_results.append(ToolCallResult(
+                    tool_name=tool,
+                    expected=True,
+                    actual=False,
+                    passed=False,
+                    metadata={"error": str(e)}
+                ))
+        
+        self.after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
+    
+    async def run_async(self, print_summary: bool = False) -> ReliabilityResult:
+        """
+        Execute the reliability evaluation asynchronously.
+        
+        Args:
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            ReliabilityResult with tool call verification results
+        """
+        await self.async_before_run()
+        result = self.run(print_summary=False)
+        await self.async_after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
+    
+    def evaluate_tool_calls(
+        self,
+        actual_tools: List[str],
+        print_summary: bool = False
+    ) -> ReliabilityResult:
+        """
+        Evaluate a pre-recorded list of tool calls without running the agent.
+        
+        Args:
+            actual_tools: List of tool names that were called
+            print_summary: Whether to print summary after evaluation
+            
+        Returns:
+            ReliabilityResult with tool call verification results
+        """
+        self.before_run()
+        
+        result = ReliabilityResult(
+            eval_id=self.eval_id,
+            name=self.name
+        )
+        
+        actual_set = set(actual_tools)
+        
+        for tool in self.expected_tools:
+            was_called = tool in actual_set
+            result.tool_results.append(ToolCallResult(
+                tool_name=tool,
+                expected=True,
+                actual=was_called,
+                passed=was_called
+            ))
+        
+        for tool in self.forbidden_tools:
+            was_called = tool in actual_set
+            result.tool_results.append(ToolCallResult(
+                tool_name=tool,
+                expected=False,
+                actual=was_called,
+                passed=not was_called
+            ))
+        
+        self.after_run(result)
+        
+        if print_summary:
+            result.print_summary()
+        
+        return result
@@ -0,0 +1,463 @@
+"""
+Result dataclasses for PraisonAI Agents evaluation framework.
+
+This module provides structured result types for different evaluation methods.
+"""
+
+from dataclasses import dataclass, field, asdict
+from typing import List, Optional, Dict, Any
+from datetime import datetime
+import statistics
+import json
+
+
+@dataclass
+class EvaluationScore:
+    """Individual evaluation score with reasoning."""
+    score: float
+    reasoning: str
+    input_text: str
+    output_text: str
+    expected_output: Optional[str] = None
+    timestamp: str = field(default_factory=lambda: datetime.now().isoformat())
+    metadata: Dict[str, Any] = field(default_factory=dict)
+
+
+@dataclass
+class AccuracyResult:
+    """Result from accuracy evaluation comparing output to expected output."""
+    evaluations: List[EvaluationScore] = field(default_factory=list)
+    eval_id: str = ""
+    name: str = ""
+    
+    @property
+    def scores(self) -> List[float]:
+        """Get list of all scores."""
+        return [e.score for e in self.evaluations]
+    
+    @property
+    def avg_score(self) -> float:
+        """Calculate average score."""
+        if not self.scores:
+            return 0.0
+        return statistics.mean(self.scores)
+    
+    @property
+    def min_score(self) -> float:
+        """Get minimum score."""
+        if not self.scores:
+            return 0.0
+        return min(self.scores)
+    
+    @property
+    def max_score(self) -> float:
+        """Get maximum score."""
+        if not self.scores:
+            return 0.0
+        return max(self.scores)
+    
+    @property
+    def std_dev(self) -> float:
+        """Calculate standard deviation of scores."""
+        if len(self.scores) < 2:
+            return 0.0
+        return statistics.stdev(self.scores)
+    
+    @property
+    def passed(self) -> bool:
+        """Check if evaluation passed (avg score >= 7)."""
+        return self.avg_score >= 7.0
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert result to dictionary."""
+        return {
+            "eval_id": self.eval_id,
+            "name": self.name,
+            "evaluations": [asdict(e) for e in self.evaluations],
+            "avg_score": self.avg_score,
+            "min_score": self.min_score,
+            "max_score": self.max_score,
+            "std_dev": self.std_dev,
+            "passed": self.passed,
+            "num_evaluations": len(self.evaluations)
+        }
+    
+    def to_json(self) -> str:
+        """Convert result to JSON string."""
+        return json.dumps(self.to_dict(), indent=2)
+    
+    def print_summary(self) -> None:
+        """Print a summary of the evaluation results."""
+        try:
+            from rich.console import Console
+            from rich.table import Table
+            from rich.panel import Panel  # noqa: F401
+            
+            console = Console()
+            
+            table = Table(title="Accuracy Evaluation Summary")
+            table.add_column("Metric", style="cyan")
+            table.add_column("Value", style="green")
+            
+            table.add_row("Evaluations", str(len(self.evaluations)))
+            table.add_row("Average Score", f"{self.avg_score:.2f}")
+            table.add_row("Min Score", f"{self.min_score:.2f}")
+            table.add_row("Max Score", f"{self.max_score:.2f}")
+            table.add_row("Std Dev", f"{self.std_dev:.2f}")
+            table.add_row("Status", "✅ PASSED" if self.passed else "❌ FAILED")
+            
+            console.print(table)
+        except ImportError:
+            print("Accuracy Evaluation Summary")
+            print(f"  Evaluations: {len(self.evaluations)}")
+            print(f"  Average Score: {self.avg_score:.2f}")
+            print(f"  Min Score: {self.min_score:.2f}")
+            print(f"  Max Score: {self.max_score:.2f}")
+            print(f"  Std Dev: {self.std_dev:.2f}")
+            print(f"  Status: {'PASSED' if self.passed else 'FAILED'}")
+
+
+@dataclass
+class PerformanceMetrics:
+    """Individual performance measurement."""
+    run_time: float
+    memory_usage: float
+    iteration: int
+    timestamp: str = field(default_factory=lambda: datetime.now().isoformat())
+    metadata: Dict[str, Any] = field(default_factory=dict)
+
+
+@dataclass
+class PerformanceResult:
+    """Result from performance evaluation measuring runtime and memory."""
+    metrics: List[PerformanceMetrics] = field(default_factory=list)
+    warmup_runs: int = 0
+    eval_id: str = ""
+    name: str = ""
+    
+    @property
+    def run_times(self) -> List[float]:
+        """Get list of all run times."""
+        return [m.run_time for m in self.metrics]
+    
+    @property
+    def memory_usages(self) -> List[float]:
+        """Get list of all memory usages."""
+        return [m.memory_usage for m in self.metrics]
+    
+    @property
+    def avg_run_time(self) -> float:
+        """Calculate average run time."""
+        if not self.run_times:
+            return 0.0
+        return statistics.mean(self.run_times)
+    
+    @property
+    def min_run_time(self) -> float:
+        """Get minimum run time."""
+        if not self.run_times:
+            return 0.0
+        return min(self.run_times)
+    
+    @property
+    def max_run_time(self) -> float:
+        """Get maximum run time."""
+        if not self.run_times:
+            return 0.0
+        return max(self.run_times)
+    
+    @property
+    def std_dev_run_time(self) -> float:
+        """Calculate standard deviation of run times."""
+        if len(self.run_times) < 2:
+            return 0.0
+        return statistics.stdev(self.run_times)
+    
+    @property
+    def median_run_time(self) -> float:
+        """Calculate median run time."""
+        if not self.run_times:
+            return 0.0
+        return statistics.median(self.run_times)
+    
+    @property
+    def p95_run_time(self) -> float:
+        """Calculate 95th percentile run time."""
+        if not self.run_times:
+            return 0.0
+        sorted_times = sorted(self.run_times)
+        idx = int(len(sorted_times) * 0.95)
+        return sorted_times[min(idx, len(sorted_times) - 1)]
+    
+    @property
+    def avg_memory(self) -> float:
+        """Calculate average memory usage."""
+        if not self.memory_usages:
+            return 0.0
+        return statistics.mean(self.memory_usages)
+    
+    @property
+    def max_memory(self) -> float:
+        """Get maximum memory usage."""
+        if not self.memory_usages:
+            return 0.0
+        return max(self.memory_usages)
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert result to dictionary."""
+        return {
+            "eval_id": self.eval_id,
+            "name": self.name,
+            "metrics": [asdict(m) for m in self.metrics],
+            "warmup_runs": self.warmup_runs,
+            "avg_run_time": self.avg_run_time,
+            "min_run_time": self.min_run_time,
+            "max_run_time": self.max_run_time,
+            "std_dev_run_time": self.std_dev_run_time,
+            "median_run_time": self.median_run_time,
+            "p95_run_time": self.p95_run_time,
+            "avg_memory": self.avg_memory,
+            "max_memory": self.max_memory,
+            "num_iterations": len(self.metrics)
+        }
+    
+    def to_json(self) -> str:
+        """Convert result to JSON string."""
+        return json.dumps(self.to_dict(), indent=2)
+    
+    def print_summary(self) -> None:
+        """Print a summary of the performance results."""
+        try:
+            from rich.console import Console
+            from rich.table import Table
+            
+            console = Console()
+            
+            table = Table(title="Performance Evaluation Summary")
+            table.add_column("Metric", style="cyan")
+            table.add_column("Value", style="green")
+            
+            table.add_row("Iterations", str(len(self.metrics)))
+            table.add_row("Warmup Runs", str(self.warmup_runs))
+            table.add_row("Avg Run Time", f"{self.avg_run_time:.4f}s")
+            table.add_row("Min Run Time", f"{self.min_run_time:.4f}s")
+            table.add_row("Max Run Time", f"{self.max_run_time:.4f}s")
+            table.add_row("Std Dev", f"{self.std_dev_run_time:.4f}s")
+            table.add_row("Median", f"{self.median_run_time:.4f}s")
+            table.add_row("P95", f"{self.p95_run_time:.4f}s")
+            table.add_row("Avg Memory", f"{self.avg_memory:.2f} MB")
+            table.add_row("Max Memory", f"{self.max_memory:.2f} MB")
+            
+            console.print(table)
+        except ImportError:
+            print("Performance Evaluation Summary")
+            print(f"  Iterations: {len(self.metrics)}")
+            print(f"  Warmup Runs: {self.warmup_runs}")
+            print(f"  Avg Run Time: {self.avg_run_time:.4f}s")
+            print(f"  Min Run Time: {self.min_run_time:.4f}s")
+            print(f"  Max Run Time: {self.max_run_time:.4f}s")
+            print(f"  Std Dev: {self.std_dev_run_time:.4f}s")
+            print(f"  Median: {self.median_run_time:.4f}s")
+            print(f"  P95: {self.p95_run_time:.4f}s")
+            print(f"  Avg Memory: {self.avg_memory:.2f} MB")
+            print(f"  Max Memory: {self.max_memory:.2f} MB")
+
+
+@dataclass
+class ToolCallResult:
+    """Result of a single tool call verification."""
+    tool_name: str
+    expected: bool
+    actual: bool
+    passed: bool
+    arguments: Optional[Dict[str, Any]] = None
+    metadata: Dict[str, Any] = field(default_factory=dict)
+
+
+@dataclass
+class ReliabilityResult:
+    """Result from reliability evaluation checking tool calls."""
+    tool_results: List[ToolCallResult] = field(default_factory=list)
+    eval_id: str = ""
+    name: str = ""
+    
+    @property
+    def passed_calls(self) -> List[ToolCallResult]:
+        """Get list of passed tool calls."""
+        return [t for t in self.tool_results if t.passed]
+    
+    @property
+    def failed_calls(self) -> List[ToolCallResult]:
+        """Get list of failed tool calls."""
+        return [t for t in self.tool_results if not t.passed]
+    
+    @property
+    def pass_rate(self) -> float:
+        """Calculate pass rate."""
+        if not self.tool_results:
+            return 0.0
+        return len(self.passed_calls) / len(self.tool_results)
+    
+    @property
+    def status(self) -> str:
+        """Get overall status."""
+        return "PASSED" if len(self.failed_calls) == 0 else "FAILED"
+    
+    @property
+    def passed(self) -> bool:
+        """Check if all tool calls passed."""
+        return len(self.failed_calls) == 0
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert result to dictionary."""
+        return {
+            "eval_id": self.eval_id,
+            "name": self.name,
+            "tool_results": [asdict(t) for t in self.tool_results],
+            "passed_count": len(self.passed_calls),
+            "failed_count": len(self.failed_calls),
+            "pass_rate": self.pass_rate,
+            "status": self.status
+        }
+    
+    def to_json(self) -> str:
+        """Convert result to JSON string."""
+        return json.dumps(self.to_dict(), indent=2)
+    
+    def print_summary(self) -> None:
+        """Print a summary of the reliability results."""
+        try:
+            from rich.console import Console
+            from rich.table import Table
+            
+            console = Console()
+            
+            table = Table(title="Reliability Evaluation Summary")
+            table.add_column("Metric", style="cyan")
+            table.add_column("Value", style="green")
+            
+            table.add_row("Total Tool Calls", str(len(self.tool_results)))
+            table.add_row("Passed", str(len(self.passed_calls)))
+            table.add_row("Failed", str(len(self.failed_calls)))
+            table.add_row("Pass Rate", f"{self.pass_rate:.1%}")
+            table.add_row("Status", "✅ PASSED" if self.passed else "❌ FAILED")
+            
+            console.print(table)
+            
+            if self.failed_calls:
+                fail_table = Table(title="Failed Tool Calls")
+                fail_table.add_column("Tool", style="red")
+                fail_table.add_column("Expected", style="yellow")
+                fail_table.add_column("Actual", style="yellow")
+                
+                for tc in self.failed_calls:
+                    fail_table.add_row(tc.tool_name, str(tc.expected), str(tc.actual))
+                
+                console.print(fail_table)
+        except ImportError:
+            print("Reliability Evaluation Summary")
+            print(f"  Total Tool Calls: {len(self.tool_results)}")
+            print(f"  Passed: {len(self.passed_calls)}")
+            print(f"  Failed: {len(self.failed_calls)}")
+            print(f"  Pass Rate: {self.pass_rate:.1%}")
+            print(f"  Status: {self.status}")
+
+
+@dataclass
+class CriteriaScore:
+    """Individual criteria evaluation score."""
+    score: float
+    passed: bool
+    reasoning: str
+    output_text: str
+    criteria: str
+    timestamp: str = field(default_factory=lambda: datetime.now().isoformat())
+    metadata: Dict[str, Any] = field(default_factory=dict)
+
+
+@dataclass
+class CriteriaResult:
+    """Result from criteria-based evaluation (agent-as-judge)."""
+    evaluations: List[CriteriaScore] = field(default_factory=list)
+    criteria: str = ""
+    scoring_type: str = "numeric"  # "numeric" or "binary"
+    threshold: float = 7.0
+    eval_id: str = ""
+    name: str = ""
+    
+    @property
+    def scores(self) -> List[float]:
+        """Get list of all scores."""
+        return [e.score for e in self.evaluations]
+    
+    @property
+    def avg_score(self) -> float:
+        """Calculate average score."""
+        if not self.scores:
+            return 0.0
+        return statistics.mean(self.scores)
+    
+    @property
+    def pass_rate(self) -> float:
+        """Calculate pass rate."""
+        if not self.evaluations:
+            return 0.0
+        passed = sum(1 for e in self.evaluations if e.passed)
+        return passed / len(self.evaluations)
+    
+    @property
+    def passed(self) -> bool:
+        """Check if evaluation passed overall."""
+        if self.scoring_type == "binary":
+            return self.pass_rate >= 0.5
+        return self.avg_score >= self.threshold
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert result to dictionary."""
+        return {
+            "eval_id": self.eval_id,
+            "name": self.name,
+            "criteria": self.criteria,
+            "scoring_type": self.scoring_type,
+            "threshold": self.threshold,
+            "evaluations": [asdict(e) for e in self.evaluations],
+            "avg_score": self.avg_score,
+            "pass_rate": self.pass_rate,
+            "passed": self.passed,
+            "num_evaluations": len(self.evaluations)
+        }
+    
+    def to_json(self) -> str:
+        """Convert result to JSON string."""
+        return json.dumps(self.to_dict(), indent=2)
+    
+    def print_summary(self) -> None:
+        """Print a summary of the criteria evaluation results."""
+        try:
+            from rich.console import Console
+            from rich.table import Table
+            
+            console = Console()
+            
+            table = Table(title="Criteria Evaluation Summary")
+            table.add_column("Metric", style="cyan")
+            table.add_column("Value", style="green")
+            
+            table.add_row("Criteria", self.criteria[:50] + "..." if len(self.criteria) > 50 else self.criteria)
+            table.add_row("Scoring Type", self.scoring_type)
+            table.add_row("Threshold", str(self.threshold))
+            table.add_row("Evaluations", str(len(self.evaluations)))
+            table.add_row("Average Score", f"{self.avg_score:.2f}")
+            table.add_row("Pass Rate", f"{self.pass_rate:.1%}")
+            table.add_row("Status", "✅ PASSED" if self.passed else "❌ FAILED")
+            
+            console.print(table)
+        except ImportError:
+            print("Criteria Evaluation Summary")
+            print(f"  Criteria: {self.criteria[:50]}...")
+            print(f"  Scoring Type: {self.scoring_type}")
+            print(f"  Threshold: {self.threshold}")
+            print(f"  Evaluations: {len(self.evaluations)}")
+            print(f"  Average Score: {self.avg_score:.2f}")
+            print(f"  Pass Rate: {self.pass_rate:.1%}")
+            print(f"  Status: {'PASSED' if self.passed else 'FAILED'}")
@@ -0,0 +1,154 @@
+"""
+Utility functions for PraisonAI Agents evaluation framework.
+"""
+
+import json
+import logging
+from pathlib import Path
+from typing import Any, Dict, Optional, Union
+
+logger = logging.getLogger(__name__)
+
+
+def save_result_to_file(
+    result: Any,
+    file_path: str,
+    eval_id: Optional[str] = None,
+    name: Optional[str] = None
+) -> bool:
+    """
+    Save evaluation result to a file.
+    
+    Args:
+        result: The evaluation result to save
+        file_path: Path to save the result (supports {name}, {eval_id} placeholders)
+        eval_id: Optional eval ID for placeholder substitution
+        name: Optional name for placeholder substitution
+        
+    Returns:
+        True if save was successful, False otherwise
+    """
+    try:
+        formatted_path = file_path.format(name=name or "", eval_id=eval_id or "")
+        path = Path(formatted_path)
+        path.parent.mkdir(parents=True, exist_ok=True)
+        
+        if hasattr(result, 'to_dict'):
+            data = result.to_dict()
+        elif hasattr(result, '__dict__'):
+            data = result.__dict__
+        else:
+            data = {"result": str(result)}
+        
+        path.write_text(json.dumps(data, indent=2, default=str))
+        logger.debug(f"Saved result to: {formatted_path}")
+        return True
+    except Exception as e:
+        logger.warning(f"Failed to save result: {e}")
+        return False
+
+
+def load_result_from_file(file_path: str) -> Optional[Dict[str, Any]]:
+    """
+    Load evaluation result from a file.
+    
+    Args:
+        file_path: Path to the result file
+        
+    Returns:
+        Dictionary containing the result data, or None if load failed
+    """
+    try:
+        path = Path(file_path)
+        if not path.exists():
+            logger.warning(f"Result file not found: {file_path}")
+            return None
+        
+        data = json.loads(path.read_text())
+        return data
+    except Exception as e:
+        logger.warning(f"Failed to load result: {e}")
+        return None
+
+
+def format_score(score: float, max_score: float = 10.0) -> str:
+    """
+    Format a score for display.
+    
+    Args:
+        score: The score value
+        max_score: Maximum possible score
+        
+    Returns:
+        Formatted score string
+    """
+    return f"{score:.1f}/{max_score:.0f}"
+
+
+def format_percentage(value: float) -> str:
+    """
+    Format a value as a percentage.
+    
+    Args:
+        value: Value between 0 and 1
+        
+    Returns:
+        Formatted percentage string
+    """
+    return f"{value * 100:.1f}%"
+
+
+def format_duration(seconds: float) -> str:
+    """
+    Format a duration in seconds for display.
+    
+    Args:
+        seconds: Duration in seconds
+        
+    Returns:
+        Formatted duration string
+    """
+    if seconds < 0.001:
+        return f"{seconds * 1000000:.0f}μs"
+    elif seconds < 1:
+        return f"{seconds * 1000:.2f}ms"
+    elif seconds < 60:
+        return f"{seconds:.2f}s"
+    else:
+        minutes = int(seconds // 60)
+        secs = seconds % 60
+        return f"{minutes}m {secs:.1f}s"
+
+
+def format_memory(mb: float) -> str:
+    """
+    Format memory usage for display.
+    
+    Args:
+        mb: Memory in megabytes
+        
+    Returns:
+        Formatted memory string
+    """
+    if mb < 1:
+        return f"{mb * 1024:.1f} KB"
+    elif mb < 1024:
+        return f"{mb:.2f} MB"
+    else:
+        return f"{mb / 1024:.2f} GB"
+
+
+def calculate_pass_rate(passed: int, total: int) -> float:
+    """
+    Calculate pass rate.
+    
+    Args:
+        passed: Number of passed evaluations
+        total: Total number of evaluations
+        
+    Returns:
+        Pass rate as a float between 0 and 1
+    """
+    if total == 0:
+        return 0.0
+    return passed / total
@@ -1054,7 +1054,7 @@ class LLM:
        **kwargs
    ) -> str:
        """Enhanced get_response with all OpenAI-like features"""
-        logging.info(f"Getting response from {self.model}")
+        logging.debug(f"Getting response from {self.model}")
        # Log all self values when in debug mode
        self._log_llm_config(
            'LLM instance',
@@ -2472,7 +2472,7 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
        """Async version of get_response with identical functionality."""
        try:
            import litellm
-            logging.info(f"Getting async response from {self.model}")
+            logging.debug(f"Getting async response from {self.model}")
            # Log all self values when in debug mode
            self._log_llm_config(
                'get_response_async',
@@ -19,6 +19,15 @@ from rich.console import Console
 from rich.live import Live
 import inspect

+# Import display_tool_call for callback support (lazy import to avoid circular imports)
+_display_tool_call = None
+def _get_display_tool_call():
+    global _display_tool_call
+    if _display_tool_call is None:
+        from ..main import display_tool_call
+        _display_tool_call = display_tool_call
+    return _display_tool_call
+
 # Constants
 LOCAL_SERVER_API_KEY_PLACEHOLDER = "not-needed"

@@ -978,16 +987,19 @@ class OpenAIClient:
                        function_name = tool_call.function.name
                        arguments = json.loads(tool_call.function.arguments)
                    
+                    # Always trigger callback for tool call tracking (even when verbose=False)
+                    display_tool_call_fn = _get_display_tool_call()
+                    display_tool_call_fn(f"Calling function: {function_name}", console=console if verbose else None)
+                    
                    if verbose and console:
-                        console.print(f"[bold]Calling function:[/bold] {function_name}")
                        console.print(f"[dim]Arguments:[/dim] {arguments}")
                    
                    # Execute the tool
                    tool_result = execute_tool_fn(function_name, arguments)
                    results_str = json.dumps(tool_result) if tool_result else "Function returned an empty output"
                    
-                    if verbose and console:
-                        console.print(f"[dim]Result:[/dim] {results_str}")
+                    # Trigger callback with result
+                    display_tool_call_fn(f"Function {function_name} returned: {results_str[:200]}{'...' if len(results_str) > 200 else ''}", console=console if verbose else None)
                    
                    messages.append({
                        "role": "tool",
@@ -1150,8 +1162,11 @@ class OpenAIClient:
                        function_name = tool_call.function.name
                        arguments = json.loads(tool_call.function.arguments)
                    
+                    # Always trigger callback for tool call tracking (even when verbose=False)
+                    display_tool_call_fn = _get_display_tool_call()
+                    display_tool_call_fn(f"Calling function: {function_name}", console=console if verbose else None)
+                    
                    if verbose and console:
-                        console.print(f"[bold]Calling function:[/bold] {function_name}")
                        console.print(f"[dim]Arguments:[/dim] {arguments}")
                    
                    # Execute the tool (async)
@@ -1167,8 +1182,8 @@ class OpenAIClient:
                    
                    results_str = json.dumps(tool_result) if tool_result else "Function returned an empty output"
                    
-                    if verbose and console:
-                        console.print(f"[dim]Result:[/dim] {results_str}")
+                    # Trigger callback with result
+                    display_tool_call_fn(f"Function {function_name} returned: {results_str[:200]}{'...' if len(results_str) > 200 else ''}", console=console if verbose else None)
                    
                    messages.append({
                        "role": "tool",
@@ -1299,6 +1314,10 @@ class OpenAIClient:
                                yield f"\n[Error parsing arguments for {function_name if 'function_name' in locals() else 'unknown function'}: {str(e)}]"
                            continue
                        
+                        # Always trigger callback for tool call tracking (even when verbose=False)
+                        display_tool_call_fn = _get_display_tool_call()
+                        display_tool_call_fn(f"Calling function: {function_name}", console=None)
+                        
                        if verbose:
                            yield f"\n[Calling function: {function_name}]"
                        
@@ -1311,6 +1330,9 @@ class OpenAIClient:
                            if verbose:
                                yield f"\n[Function error: {str(e)}]"
                        
+                        # Trigger callback with result
+                        display_tool_call_fn(f"Function {function_name} returned: {results_str[:200]}{'...' if len(results_str) > 200 else ''}", console=None)
+                        
                        if verbose:
                            yield f"\n[Function result: {results_str}]"
                        
@@ -231,15 +231,15 @@ def display_tool_call(message: str, console=None):
    if not message or not message.strip():
        logging.debug("Empty message in display_tool_call, returning early")
        return
-    if console is None:
-        console = Console()
    message = _clean_display_content(str(message))
    logging.debug(f"Cleaned message in display_tool_call: {repr(message)}")
    
-    # Execute synchronous callbacks
+    # Execute synchronous callbacks (always, even when console is None)
    execute_sync_callback('tool_call', message=message)
    
-    console.print(Panel.fit(Text(message, style="bold cyan"), title="Tool Call", border_style="green"))
+    # Only print panel if console is provided (verbose mode)
+    if console is not None:
+        console.print(Panel.fit(Text(message, style="bold cyan"), title="Tool Call", border_style="green"))

 def display_error(message: str, console=None):
    if not message or not message.strip():
@@ -0,0 +1,76 @@
+"""Agent Skills module for PraisonAI Agents.
+
+This module provides support for the open Agent Skills standard (agentskills.io),
+enabling agents to load and use modular capabilities through SKILL.md files.
+
+Zero Performance Impact:
+- All imports are lazy loaded via __getattr__
+- Skills only loaded when explicitly enabled
+- No auto-discovery at import time
+
+Usage:
+    from praisonaiagents.skills import SkillManager, SkillProperties
+    
+    manager = SkillManager()
+    manager.discover(["./skills"])
+    prompt_xml = manager.to_prompt()
+"""
+
+__all__ = [
+    # Models
+    "SkillProperties",
+    "SkillMetadata",
+    "ParseError",
+    "ValidationError",
+    # Parser
+    "parse_frontmatter",
+    "find_skill_md",
+    "read_properties",
+    # Validator
+    "validate",
+    "validate_metadata",
+    # Prompt
+    "to_prompt",
+    "generate_skills_xml",
+    "format_skill_for_prompt",
+    # Discovery
+    "discover_skills",
+    "get_default_skill_dirs",
+    # Loader
+    "SkillLoader",
+    # Manager
+    "SkillManager",
+]
+
+
+def __getattr__(name: str):
+    """Lazy load module components to avoid import overhead."""
+    if name in ("SkillProperties", "SkillMetadata", "ParseError", "ValidationError"):
+        from .models import SkillProperties, SkillMetadata, ParseError, ValidationError
+        return locals()[name]
+    
+    if name in ("parse_frontmatter", "find_skill_md", "read_properties"):
+        from .parser import parse_frontmatter, find_skill_md, read_properties
+        return locals()[name]
+    
+    if name in ("validate", "validate_metadata", "_validate_name", "_validate_description", "_validate_compatibility"):
+        from .validator import validate, validate_metadata, _validate_name, _validate_description, _validate_compatibility
+        return locals()[name]
+    
+    if name in ("to_prompt", "generate_skills_xml", "format_skill_for_prompt"):
+        from .prompt import to_prompt, generate_skills_xml, format_skill_for_prompt
+        return locals()[name]
+    
+    if name in ("discover_skills", "get_default_skill_dirs"):
+        from .discovery import discover_skills, get_default_skill_dirs
+        return locals()[name]
+    
+    if name == "SkillLoader":
+        from .loader import SkillLoader
+        return SkillLoader
+    
+    if name == "SkillManager":
+        from .manager import SkillManager
+        return SkillManager
+    
+    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
@@ -0,0 +1,129 @@
+"""Skill discovery and directory scanning."""
+
+from pathlib import Path
+from typing import List, Optional
+
+from .parser import find_skill_md, read_properties
+from .models import SkillProperties
+
+
+def get_default_skill_dirs() -> List[Path]:
+    """Get default skill directory locations.
+
+    Returns directories in precedence order (high to low):
+    1. Project: ./.praison/skills/ or ./.claude/skills/
+    2. User: ~/.praison/skills/
+    3. System: /etc/praison/skills/ (admin-managed)
+
+    Returns:
+        List of existing skill directories
+    """
+    dirs = []
+    cwd = Path.cwd()
+
+    # Project-level directories
+    project_dirs = [
+        cwd / ".praison" / "skills",
+        cwd / ".claude" / "skills",
+    ]
+    for d in project_dirs:
+        if d.exists() and d.is_dir():
+            dirs.append(d)
+
+    # User-level directory
+    user_dir = Path.home() / ".praison" / "skills"
+    if user_dir.exists() and user_dir.is_dir():
+        dirs.append(user_dir)
+
+    # System-level directory (Unix-like systems)
+    system_dir = Path("/etc/praison/skills")
+    if system_dir.exists() and system_dir.is_dir():
+        dirs.append(system_dir)
+
+    return dirs
+
+
+def discover_skills(
+    skill_dirs: Optional[List[str]] = None,
+    include_defaults: bool = True,
+) -> List[SkillProperties]:
+    """Discover all valid skills in the given directories.
+
+    Args:
+        skill_dirs: List of directory paths to scan for skills.
+            Each directory should contain skill subdirectories.
+        include_defaults: Whether to include default skill directories
+
+    Returns:
+        List of SkillProperties for all valid skills found
+    """
+    all_dirs = []
+
+    # Add explicit directories
+    if skill_dirs:
+        for d in skill_dirs:
+            path = Path(d).expanduser().resolve()
+            if path.exists() and path.is_dir():
+                all_dirs.append(path)
+
+    # Add default directories
+    if include_defaults:
+        all_dirs.extend(get_default_skill_dirs())
+
+    # Remove duplicates while preserving order
+    seen = set()
+    unique_dirs = []
+    for d in all_dirs:
+        if d not in seen:
+            seen.add(d)
+            unique_dirs.append(d)
+
+    skills = []
+
+    for parent_dir in unique_dirs:
+        # Each subdirectory in parent_dir might be a skill
+        try:
+            for item in parent_dir.iterdir():
+                if not item.is_dir():
+                    continue
+
+                # Check if this directory contains a SKILL.md
+                skill_md = find_skill_md(item)
+                if skill_md is None:
+                    continue
+
+                try:
+                    props = read_properties(item)
+                    skills.append(props)
+                except Exception:
+                    # Skip invalid skills
+                    continue
+        except PermissionError:
+            # Skip directories we can't read
+            continue
+
+    return skills
+
+
+def discover_skill(skill_path: str) -> Optional[SkillProperties]:
+    """Discover a single skill from a directory path.
+
+    Args:
+        skill_path: Path to a skill directory
+
+    Returns:
+        SkillProperties if valid, None otherwise
+    """
+    path = Path(skill_path).expanduser().resolve()
+
+    if not path.exists() or not path.is_dir():
+        return None
+
+    skill_md = find_skill_md(path)
+    if skill_md is None:
+        return None
+
+    try:
+        return read_properties(path)
+    except Exception:
+        return None
@@ -0,0 +1,236 @@
+"""Progressive skill loading for Agent Skills."""
+
+from pathlib import Path
+from typing import Optional
+from dataclasses import dataclass, field
+
+from .parser import find_skill_md, read_properties, parse_frontmatter
+from .models import SkillProperties, SkillMetadata
+
+
+@dataclass
+class LoadedSkill:
+    """A fully or partially loaded skill.
+
+    Supports progressive disclosure:
+    - Level 1: metadata only (name, description, location)
+    - Level 2: full instructions (SKILL.md body)
+    - Level 3: resources (scripts, references, assets)
+    """
+
+    properties: SkillProperties
+    instructions: Optional[str] = None
+    resources_loaded: bool = False
+    _scripts: dict = field(default_factory=dict)
+    _references: dict = field(default_factory=dict)
+    _assets: dict = field(default_factory=dict)
+
+    @property
+    def metadata(self) -> SkillMetadata:
+        """Get lightweight metadata for system prompt."""
+        return SkillMetadata.from_properties(self.properties)
+
+    @property
+    def is_activated(self) -> bool:
+        """Check if skill has been activated (instructions loaded)."""
+        return self.instructions is not None
+
+    def get_scripts(self) -> dict:
+        """Get loaded scripts."""
+        return self._scripts
+
+    def get_references(self) -> dict:
+        """Get loaded references."""
+        return self._references
+
+    def get_assets(self) -> dict:
+        """Get loaded assets."""
+        return self._assets
+
+
+class SkillLoader:
+    """Loader for Agent Skills with progressive disclosure support.
+
+    Usage:
+        loader = SkillLoader()
+
+        # Level 1: Load metadata only
+        skill = loader.load_metadata("/path/to/skill")
+
+        # Level 2: Activate skill (load instructions)
+        loader.activate(skill)
+
+        # Level 3: Load specific resources
+        loader.load_scripts(skill)
+        loader.load_references(skill)
+    """
+
+    def load_metadata(self, skill_path: str) -> Optional[LoadedSkill]:
+        """Load skill metadata only (Level 1).
+
+        This is the minimal load for system prompt injection.
+        Only reads the YAML frontmatter, not the full SKILL.md body.
+
+        Args:
+            skill_path: Path to skill directory
+
+        Returns:
+            LoadedSkill with metadata only, or None if invalid
+        """
+        path = Path(skill_path).expanduser().resolve()
+
+        if not path.exists() or not path.is_dir():
+            return None
+
+        try:
+            props = read_properties(path)
+            return LoadedSkill(properties=props)
+        except Exception:
+            return None
+
+    def activate(self, skill: LoadedSkill) -> bool:
+        """Activate a skill by loading its full instructions (Level 2).
+
+        Args:
+            skill: LoadedSkill instance to activate
+
+        Returns:
+            True if activation succeeded
+        """
+        if skill.is_activated:
+            return True
+
+        if skill.properties.path is None:
+            return False
+
+        skill_md = find_skill_md(skill.properties.path)
+        if skill_md is None:
+            return False
+
+        try:
+            content = skill_md.read_text()
+            _, body = parse_frontmatter(content)
+            skill.instructions = body
+            return True
+        except Exception:
+            return False
+
+    def load_scripts(self, skill: LoadedSkill) -> dict:
+        """Load scripts from skill's scripts/ directory (Level 3).
+
+        Args:
+            skill: LoadedSkill instance
+
+        Returns:
+            Dict mapping script names to their contents
+        """
+        if skill.properties.path is None:
+            return {}
+
+        scripts_dir = skill.properties.path / "scripts"
+        if not scripts_dir.exists() or not scripts_dir.is_dir():
+            return {}
+
+        scripts = {}
+        try:
+            for script_file in scripts_dir.iterdir():
+                if script_file.is_file():
+                    try:
+                        scripts[script_file.name] = script_file.read_text()
+                    except Exception:
+                        continue
+        except PermissionError:
+            pass
+
+        skill._scripts = scripts
+        return scripts
+
+    def load_references(self, skill: LoadedSkill) -> dict:
+        """Load references from skill's references/ directory (Level 3).
+
+        Args:
+            skill: LoadedSkill instance
+
+        Returns:
+            Dict mapping reference names to their contents
+        """
+        if skill.properties.path is None:
+            return {}
+
+        refs_dir = skill.properties.path / "references"
+        if not refs_dir.exists() or not refs_dir.is_dir():
+            return {}
+
+        refs = {}
+        try:
+            for ref_file in refs_dir.iterdir():
+                if ref_file.is_file():
+                    try:
+                        refs[ref_file.name] = ref_file.read_text()
+                    except Exception:
+                        continue
+        except PermissionError:
+            pass
+
+        skill._references = refs
+        return refs
+
+    def load_assets(self, skill: LoadedSkill) -> dict:
+        """Load asset paths from skill's assets/ directory (Level 3).
+
+        Note: For binary assets, only paths are returned, not contents.
+
+        Args:
+            skill: LoadedSkill instance
+
+        Returns:
+            Dict mapping asset names to their paths
+        """
+        if skill.properties.path is None:
+            return {}
+
+        assets_dir = skill.properties.path / "assets"
+        if not assets_dir.exists() or not assets_dir.is_dir():
+            return {}
+
+        assets = {}
+        try:
+            for asset_file in assets_dir.iterdir():
+                if asset_file.is_file():
+                    assets[asset_file.name] = str(asset_file)
+        except PermissionError:
+            pass
+
+        skill._assets = assets
+        skill.resources_loaded = True
+        return assets
+
+    def load_all_resources(self, skill: LoadedSkill) -> None:
+        """Load all resources for a skill (Level 3 complete).
+
+        Args:
+            skill: LoadedSkill instance
+        """
+        self.load_scripts(skill)
+        self.load_references(skill)
+        self.load_assets(skill)
+        skill.resources_loaded = True
+
+    @classmethod
+    def load(cls, skill_path: str, activate: bool = False) -> Optional[LoadedSkill]:
+        """Convenience method to load a skill.
+
+        Args:
+            skill_path: Path to skill directory
+            activate: Whether to also load instructions
+
+        Returns:
+            LoadedSkill or None if invalid
+        """
+        loader = cls()
+        skill = loader.load_metadata(skill_path)
+
+        if skill and activate:
+            loader.activate(skill)
+
+        return skill
@@ -0,0 +1,189 @@
+"""SkillManager for Agent Skills integration."""
+
+from typing import List, Optional, Dict
+
+from .models import SkillMetadata
+from .discovery import discover_skills
+from .loader import SkillLoader, LoadedSkill
+from .prompt import generate_skills_xml
+
+
+class SkillManager:
+    """Manager for discovering, loading, and using Agent Skills.
+
+    This is the main entry point for Agent Skills integration.
+    Supports progressive disclosure and zero performance impact
+    when skills are not used.
+
+    Usage:
+        manager = SkillManager()
+
+        # Discover skills from directories
+        manager.discover(["./skills", "~/.praison/skills"])
+
+        # Get XML for system prompt
+        prompt_xml = manager.to_prompt()
+
+        # Activate a specific skill
+        skill = manager.get_skill("pdf-processing")
+        manager.activate(skill)
+    """
+
+    def __init__(self):
+        """Initialize the SkillManager."""
+        self._skills: Dict[str, LoadedSkill] = {}
+        self._loader = SkillLoader()
+        self._discovered = False
+
+    @property
+    def skills(self) -> List[LoadedSkill]:
+        """Get all loaded skills."""
+        return list(self._skills.values())
+
+    @property
+    def skill_names(self) -> List[str]:
+        """Get names of all loaded skills."""
+        return list(self._skills.keys())
+
+    def discover(
+        self,
+        skill_dirs: Optional[List[str]] = None,
+        include_defaults: bool = True,
+    ) -> int:
+        """Discover skills from directories.
+
+        Args:
+            skill_dirs: List of directory paths to scan
+            include_defaults: Whether to include default skill directories
+
+        Returns:
+            Number of skills discovered
+        """
+        props_list = discover_skills(skill_dirs, include_defaults)
+
+        for props in props_list:
+            if props.name not in self._skills:
+                self._skills[props.name] = LoadedSkill(properties=props)
+
+        self._discovered = True
+        return len(props_list)
+
+    def add_skill(self, skill_path: str) -> Optional[LoadedSkill]:
+        """Add a single skill from a directory path.
+
+        Args:
+            skill_path: Path to skill directory
+
+        Returns:
+            LoadedSkill if successful, None otherwise
+        """
+        skill = self._loader.load_metadata(skill_path)
+        if skill:
+            self._skills[skill.properties.name] = skill
+        return skill
+
+    def get_skill(self, name: str) -> Optional[LoadedSkill]:
+        """Get a skill by name.
+
+        Args:
+            name: Skill name
+
+        Returns:
+            LoadedSkill or None if not found
+        """
+        return self._skills.get(name)
+
+    def activate(self, skill: LoadedSkill) -> bool:
+        """Activate a skill (load its instructions).
+
+        Args:
+            skill: LoadedSkill to activate
+
+        Returns:
+            True if activation succeeded
+        """
+        return self._loader.activate(skill)
+
+    def activate_by_name(self, name: str) -> bool:
+        """Activate a skill by name.
+
+        Args:
+            name: Skill name
+
+        Returns:
+            True if activation succeeded
+        """
+        skill = self.get_skill(name)
+        if skill is None:
+            return False
+        return self.activate(skill)
+
+    def get_available_skills(self) -> List[SkillMetadata]:
+        """Get metadata for all available skills.
+
+        This is used for system prompt injection.
+
+        Returns:
+            List of SkillMetadata instances
+        """
+        return [skill.metadata for skill in self._skills.values()]
+
+    def to_prompt(self) -> str:
+        """Generate XML prompt for available skills.
+
+        Returns:
+            XML string with <available_skills> block
+        """
+        metadata_list = self.get_available_skills()
+        return generate_skills_xml(metadata_list)
+
+    def get_instructions(self, name: str) -> Optional[str]:
+        """Get instructions for a skill, activating if needed.
+
+        Args:
+            name: Skill name
+
+        Returns:
+            Skill instructions or None if not found
+        """
+        skill = self.get_skill(name)
+        if skill is None:
+            return None
+
+        if not skill.is_activated:
+            self.activate(skill)
+
+        return skill.instructions
+
+    def load_resources(self, name: str) -> bool:
+        """Load all resources for a skill.
+
+        Args:
+            name: Skill name
+
+        Returns:
+            True if resources loaded successfully
+        """
+        skill = self.get_skill(name)
+        if skill is None:
+            return False
+
+        self._loader.load_all_resources(skill)
+        return True
+
+    def clear(self) -> None:
+        """Clear all loaded skills."""
+        self._skills.clear()
+        self._discovered = False
+
+    def __len__(self) -> int:
+        """Get number of loaded skills."""
+        return len(self._skills)
+
+    def __contains__(self, name: str) -> bool:
+        """Check if a skill is loaded."""
+        return name in self._skills
+
+    def __iter__(self):
+        """Iterate over loaded skills."""
+        return iter(self._skills.values())
@@ -0,0 +1,95 @@
+"""Data models for Agent Skills."""
+
+from dataclasses import dataclass, field
+from typing import Optional
+from pathlib import Path
+
+
+class ParseError(Exception):
+    """Raised when SKILL.md parsing fails."""
+    pass
+
+
+class ValidationError(Exception):
+    """Raised when skill validation fails."""
+    pass
+
+
+@dataclass
+class SkillProperties:
+    """Properties parsed from a skill's SKILL.md frontmatter.
+
+    Attributes:
+        name: Skill name in kebab-case (required)
+        description: What the skill does and when the model should use it (required)
+        license: License for the skill (optional)
+        compatibility: Compatibility information for the skill (optional)
+        allowed_tools: Tool patterns the skill requires (optional, experimental)
+        metadata: Key-value pairs for client-specific properties (defaults to
+            empty dict; omitted from to_dict() output when empty)
+        path: Path to the skill directory (optional, added for PraisonAI)
+    """
+
+    name: str
+    description: str
+    license: Optional[str] = None
+    compatibility: Optional[str] = None
+    allowed_tools: Optional[str] = None
+    metadata: dict = field(default_factory=dict)
+    path: Optional[Path] = None
+
+    def to_dict(self) -> dict:
+        """Convert to dictionary, excluding None values and empty metadata."""
+        result = {"name": self.name, "description": self.description}
+        if self.license is not None:
+            result["license"] = self.license
+        if self.compatibility is not None:
+            result["compatibility"] = self.compatibility
+        if self.allowed_tools is not None:
+            result["allowed-tools"] = self.allowed_tools
+        if self.metadata:
+            result["metadata"] = self.metadata
+        return result
+
+
+@dataclass
+class SkillMetadata:
+    """Lightweight skill metadata for system prompt injection.
+    
+    This is the minimal information needed to include a skill in the
+    system prompt's <available_skills> block (~50-100 tokens per skill).
+
+    Attributes:
+        name: Skill name
+        description: What the skill does and when to use it
+        location: Absolute path to the SKILL.md file
+    """
+
+    name: str
+    description: str
+    location: str
+
+    @classmethod
+    def from_properties(cls, props: SkillProperties) -> "SkillMetadata":
+        """Create SkillMetadata from SkillProperties.
+        
+        Args:
+            props: SkillProperties instance with path set
+            
+        Returns:
+            SkillMetadata with location pointing to SKILL.md
+        """
+        if props.path:
+            # Look for SKILL.md or skill.md
+            skill_md = props.path / "SKILL.md"
+            if not skill_md.exists():
+                skill_md = props.path / "skill.md"
+            location = str(skill_md)
+        else:
+            location = ""
+        
+        return cls(
+            name=props.name,
+            description=props.description,
+            location=location
+        )
@@ -0,0 +1,165 @@
+"""YAML frontmatter parsing for SKILL.md files."""
+
+from pathlib import Path
+from typing import Optional, Tuple
+import re
+
+from .models import ParseError, ValidationError, SkillProperties
+
+
+def find_skill_md(skill_dir: Path) -> Optional[Path]:
+    """Find the SKILL.md file in a skill directory.
+
+    Prefers SKILL.md (uppercase) but accepts skill.md (lowercase).
+
+    Args:
+        skill_dir: Path to the skill directory
+
+    Returns:
+        Path to the SKILL.md file, or None if not found
+    """
+    for name in ("SKILL.md", "skill.md"):
+        path = skill_dir / name
+        if path.exists():
+            return path
+    return None
+
+
+def parse_frontmatter(content: str) -> Tuple[dict, str]:
+    """Parse YAML frontmatter from SKILL.md content.
+
+    Args:
+        content: Raw content of SKILL.md file
+
+    Returns:
+        Tuple of (metadata dict, markdown body)
+
+    Raises:
+        ParseError: If frontmatter is missing or invalid
+    """
+    if not content.startswith("---"):
+        raise ParseError("SKILL.md must start with YAML frontmatter (---)")
+
+    parts = content.split("---", 2)
+    if len(parts) < 3:
+        raise ParseError("SKILL.md frontmatter not properly closed with ---")
+
+    frontmatter_str = parts[1]
+    body = parts[2].strip()
+
+    try:
+        metadata = _parse_yaml(frontmatter_str)
+    except Exception as e:
+        raise ParseError(f"Invalid YAML in frontmatter: {e}")
+
+    if not isinstance(metadata, dict):
+        raise ParseError("SKILL.md frontmatter must be a YAML mapping")
+
+    # Convert metadata values to strings where needed
+    if "metadata" in metadata and isinstance(metadata["metadata"], dict):
+        metadata["metadata"] = {str(k): str(v) for k, v in metadata["metadata"].items()}
+
+    return metadata, body
+
+
+def _parse_yaml(yaml_str: str) -> dict:
+    """Parse YAML string to dictionary.
+    
+    Uses a simple parser to avoid strictyaml dependency.
+    Falls back to yaml if available.
+    
+    Args:
+        yaml_str: YAML string to parse
+        
+    Returns:
+        Parsed dictionary
+    """
+    # Try to use PyYAML if available
+    try:
+        import yaml
+        return yaml.safe_load(yaml_str) or {}
+    except ImportError:
+        pass
+    
+    # Simple fallback parser for basic YAML
+    result = {}
+    current_key = None
+    current_dict = None
+    
+    for line in yaml_str.strip().split("\n"):
+        if not line.strip():
+            continue
+            
+        # Check for nested dict
+        if line.startswith("  ") and current_dict is not None:
+            match = re.match(r"^\s+(\w+):\s*(.*)$", line)
+            if match:
+                key, value = match.groups()
+                current_dict[key] = value.strip().strip('"').strip("'")
+            continue
+        
+        # Top-level key
+        match = re.match(r"^(\S+):\s*(.*)$", line)
+        if match:
+            key, value = match.groups()
+            value = value.strip()
+            
+            if not value:
+                # Could be start of nested dict
+                result[key] = {}
+                current_key = key
+                current_dict = result[key]
+            else:
+                result[key] = value.strip('"').strip("'")
+                current_dict = None
+    
+    return result
+
+
+def read_properties(skill_dir: Path) -> SkillProperties:
+    """Read skill properties from SKILL.md frontmatter.
+
+    This function parses the frontmatter and returns properties.
+    It does NOT perform full validation. Use validate() for that.
+
+    Args:
+        skill_dir: Path to the skill directory
+
+    Returns:
+        SkillProperties with parsed metadata
+
+    Raises:
+        ParseError: If SKILL.md is missing or has invalid YAML
+        ValidationError: If required fields (name, description) are missing
+    """
+    skill_dir = Path(skill_dir)
+    skill_md = find_skill_md(skill_dir)
+
+    if skill_md is None:
+        raise ParseError(f"SKILL.md not found in {skill_dir}")
+
+    content = skill_md.read_text()
+    metadata, _ = parse_frontmatter(content)
+
+    if "name" not in metadata:
+        raise ValidationError("Missing required field in frontmatter: name")
+    if "description" not in metadata:
+        raise ValidationError("Missing required field in frontmatter: description")
+
+    name = metadata["name"]
+    description = metadata["description"]
+
+    if not isinstance(name, str) or not name.strip():
+        raise ValidationError("Field 'name' must be a non-empty string")
+    if not isinstance(description, str) or not description.strip():
+        raise ValidationError("Field 'description' must be a non-empty string")
+
+    return SkillProperties(
+        name=name.strip(),
+        description=description.strip(),
+        license=metadata.get("license"),
+        compatibility=metadata.get("compatibility"),
+        allowed_tools=metadata.get("allowed-tools"),
+        metadata=metadata.get("metadata") or {},
+        path=skill_dir,
+    )
@@ -0,0 +1,74 @@
+"""XML prompt generation for Agent Skills."""
+
+from pathlib import Path
+from typing import List
+import html
+
+from .models import SkillMetadata
+from .parser import read_properties
+
+
+def format_skill_for_prompt(skill: SkillMetadata) -> str:
+    """Format a single skill as XML for system prompt.
+
+    Args:
+        skill: SkillMetadata instance
+
+    Returns:
+        XML string for the skill
+    """
+    name = html.escape(skill.name)
+    description = html.escape(skill.description)
+    location = html.escape(skill.location)
+
+    return f"""  <skill>
+    <name>{name}</name>
+    <description>{description}</description>
+    <location>{location}</location>
+  </skill>"""
+
+
+def generate_skills_xml(skills: List[SkillMetadata]) -> str:
+    """Generate XML block for available skills.
+
+    Args:
+        skills: List of SkillMetadata instances
+
+    Returns:
+        XML string with <available_skills> block
+    """
+    if not skills:
+        return "<available_skills>\n</available_skills>"
+
+    skill_entries = "\n".join(format_skill_for_prompt(s) for s in skills)
+    return f"<available_skills>\n{skill_entries}\n</available_skills>"
+
+
+def to_prompt(skill_dirs: List[Path]) -> str:
+    """Generate prompt XML from skill directories.
+
+    This is the main entry point for generating the <available_skills>
+    block to inject into agent system prompts.
+
+    Args:
+        skill_dirs: List of paths to skill directories
+
+    Returns:
+        XML string with <available_skills> block
+    """
+    skills = []
+
+    for skill_dir in skill_dirs:
+        skill_dir = Path(skill_dir)
+        if not skill_dir.exists() or not skill_dir.is_dir():
+            continue
+
+        try:
+            props = read_properties(skill_dir)
+            meta = SkillMetadata.from_properties(props)
+            skills.append(meta)
+        except Exception:
+            # Skip invalid skills silently
+            continue
+
+    return generate_skills_xml(skills)
@@ -0,0 +1,176 @@
+"""Skill validation logic."""
+
+import unicodedata
+from pathlib import Path
+from typing import Optional, List
+
+from .models import ParseError
+from .parser import find_skill_md, parse_frontmatter
+
+MAX_SKILL_NAME_LENGTH = 64
+MAX_DESCRIPTION_LENGTH = 1024
+MAX_COMPATIBILITY_LENGTH = 500
+
+ALLOWED_FIELDS = {
+    "name",
+    "description",
+    "license",
+    "allowed-tools",
+    "metadata",
+    "compatibility",
+}
+
+
+def _validate_name(name: str, skill_dir: Optional[Path]) -> List[str]:
+    """Validate skill name format and directory match.
+
+    Skill names support i18n characters (Unicode letters) plus hyphens.
+    Names must be lowercase and cannot start/end with hyphens.
+    """
+    errors = []
+
+    if not name or not isinstance(name, str) or not name.strip():
+        errors.append("Field 'name' must be a non-empty string")
+        return errors
+
+    name = unicodedata.normalize("NFKC", name.strip())
+
+    if len(name) > MAX_SKILL_NAME_LENGTH:
+        errors.append(
+            f"Skill name '{name}' exceeds {MAX_SKILL_NAME_LENGTH} character limit "
+            f"({len(name)} chars)"
+        )
+
+    if name != name.lower():
+        errors.append(f"Skill name '{name}' must be lowercase")
+
+    if name.startswith("-") or name.endswith("-"):
+        errors.append("Skill name cannot start or end with a hyphen")
+
+    if "--" in name:
+        errors.append("Skill name cannot contain consecutive hyphens")
+
+    if not all(c.isalnum() or c == "-" for c in name):
+        errors.append(
+            f"Skill name '{name}' contains invalid characters. "
+            "Only letters, digits, and hyphens are allowed."
+        )
+
+    if skill_dir:
+        dir_name = unicodedata.normalize("NFKC", skill_dir.name)
+        if dir_name != name:
+            errors.append(
+                f"Directory name '{skill_dir.name}' must match skill name '{name}'"
+            )
+
+    return errors
+
+
+def _validate_description(description: str) -> List[str]:
+    """Validate description format."""
+    errors = []
+
+    if not description or not isinstance(description, str) or not description.strip():
+        errors.append("Field 'description' must be a non-empty string")
+        return errors
+
+    if len(description) > MAX_DESCRIPTION_LENGTH:
+        errors.append(
+            f"Description exceeds {MAX_DESCRIPTION_LENGTH} character limit "
+            f"({len(description)} chars)"
+        )
+
+    return errors
+
+
+def _validate_compatibility(compatibility: str) -> List[str]:
+    """Validate compatibility format."""
+    errors = []
+
+    if not isinstance(compatibility, str):
+        errors.append("Field 'compatibility' must be a string")
+        return errors
+
+    if len(compatibility) > MAX_COMPATIBILITY_LENGTH:
+        errors.append(
+            f"Compatibility exceeds {MAX_COMPATIBILITY_LENGTH} character limit "
+            f"({len(compatibility)} chars)"
+        )
+
+    return errors
+
+
+def _validate_metadata_fields(metadata: dict) -> List[str]:
+    """Validate that only allowed fields are present."""
+    errors = []
+
+    extra_fields = set(metadata.keys()) - ALLOWED_FIELDS
+    if extra_fields:
+        errors.append(
+            f"Unexpected fields in frontmatter: {', '.join(sorted(extra_fields))}. "
+            f"Only {sorted(ALLOWED_FIELDS)} are allowed."
+        )
+
+    return errors
+
+
+def validate_metadata(metadata: dict, skill_dir: Optional[Path] = None) -> List[str]:
+    """Validate parsed skill metadata.
+
+    This is the core validation function that works on already-parsed metadata,
+    avoiding duplicate file I/O when called from the parser.
+
+    Args:
+        metadata: Parsed YAML frontmatter dictionary
+        skill_dir: Optional path to skill directory (for name-directory match check)
+
+    Returns:
+        List of validation error messages. Empty list means valid.
+    """
+    errors = []
+    errors.extend(_validate_metadata_fields(metadata))
+
+    if "name" not in metadata:
+        errors.append("Missing required field in frontmatter: name")
+    else:
+        errors.extend(_validate_name(metadata["name"], skill_dir))
+
+    if "description" not in metadata:
+        errors.append("Missing required field in frontmatter: description")
+    else:
+        errors.extend(_validate_description(metadata["description"]))
+
+    if "compatibility" in metadata:
+        errors.extend(_validate_compatibility(metadata["compatibility"]))
+
+    return errors
+
+
+def validate(skill_dir: Path) -> List[str]:
+    """Validate a skill directory.
+
+    Args:
+        skill_dir: Path to the skill directory
+
+    Returns:
+        List of validation error messages. Empty list means valid.
+    """
+    skill_dir = Path(skill_dir)
+
+    if not skill_dir.exists():
+        return [f"Path does not exist: {skill_dir}"]
+
+    if not skill_dir.is_dir():
+        return [f"Not a directory: {skill_dir}"]
+
+    skill_md = find_skill_md(skill_dir)
+    if skill_md is None:
+        return ["Missing required file: SKILL.md"]
+
+    try:
+        content = skill_md.read_text()
+        metadata, _ = parse_frontmatter(content)
+    except ParseError as e:
+        return [str(e)]
+
+    return validate_metadata(metadata, skill_dir)
@@ -218,6 +218,10 @@ TOOL_MAPPINGS = {
    'crawl4ai_extract_sync': ('.crawl4ai_tools', None),
    'Crawl4AITools': ('.crawl4ai_tools', 'Crawl4AITools'),
    'crawl4ai_tools': ('.crawl4ai_tools', None),
+    
+    # Unified Web Search (auto-fallback across providers)
+    'search_web': ('.web_search', None),
+    'get_available_providers': ('.web_search', None),
 }

 _instances = {}  # Cache for class instances
@@ -254,7 +258,8 @@ def __getattr__(name: str) -> Any:
            'exa', 'exa_search', 'exa_search_contents', 'exa_find_similar', 'exa_answer',
            'exa_search_async', 'exa_search_contents_async', 'exa_answer_async',
            'crawl4ai', 'crawl4ai_many', 'crawl4ai_extract', 'crawl4ai_llm_extract',
-            'crawl4ai_sync', 'crawl4ai_extract_sync'
+            'crawl4ai_sync', 'crawl4ai_extract_sync',
+            'search_web', 'get_available_providers'
        ]:
            return getattr(module, name)
        if name in ['file_tools', 'pandas_tools', 'wikipedia_tools',
@@ -14,6 +14,7 @@ Usage:
    context, cleaned_prompt = parser.process("@file:main.py explain this code")
 """

+import os
 import re
 import logging
 from pathlib import Path
@@ -43,10 +44,14 @@ class MentionsParser:
        "url": re.compile(r'@url:(https?://[^\s]+)'),
    }
    
+    # Default max file chars: 500K (~125K tokens) - fits GPT-4o (128K), Claude 3.5 (200K)
+    DEFAULT_MAX_FILE_CHARS = 500000
+    
    def __init__(
        self,
        workspace_path: Optional[str] = None,
-        verbose: int = 0
+        verbose: int = 0,
+        max_file_chars: Optional[int] = None
    ):
        """
        Initialize MentionsParser.
@@ -54,10 +59,24 @@ class MentionsParser:
        Args:
            workspace_path: Path to workspace/project root
            verbose: Verbosity level
+            max_file_chars: Maximum characters to include from files.
+                           Default: 500000 (500K chars, ~125K tokens).
+                           Set to 0 for no limit.
+                           Can be overridden via PRAISON_MAX_FILE_CHARS env var.
        """
        self.workspace_path = Path(workspace_path) if workspace_path else Path.cwd()
        self.verbose = verbose
        
+        # Configure max file chars: constructor > env var > default
+        if max_file_chars is not None:
+            self.max_file_chars = max_file_chars
+        else:
+            env_val = os.environ.get('PRAISON_MAX_FILE_CHARS')
+            if env_val is not None:
+                self.max_file_chars = int(env_val)
+            else:
+                self.max_file_chars = self.DEFAULT_MAX_FILE_CHARS
+        
        # Lazy-loaded managers
        self._docs_manager = None
        self._rules_manager = None
@@ -158,9 +177,15 @@ class MentionsParser:
            
            content = full_path.read_text(encoding="utf-8")
            
-            # Limit content size
-            if len(content) > 50000:
-                content = content[:50000] + "\n... (truncated)"
+            # Limit content size (0 means no limit)
+            if self.max_file_chars > 0 and len(content) > self.max_file_chars:
+                original_size = len(content)
+                content = content[:self.max_file_chars] + "\n... (truncated)"
+                self._log(
+                    f"File {file_path} truncated from {original_size:,} to {self.max_file_chars:,} chars. "
+                    f"Set PRAISON_MAX_FILE_CHARS=0 for no limit.",
+                    logging.WARNING
+                )
            
            # Detect language for code fence
            ext = full_path.suffix.lower()
@@ -253,9 +278,9 @@ class MentionsParser:
            with urllib.request.urlopen(req, timeout=10) as response:
                content = response.read().decode('utf-8', errors='ignore')
            
-            # Limit content size
-            if len(content) > 50000:
-                content = content[:50000] + "\n... (truncated)"
+            # Limit content size (0 means no limit)
+            if self.max_file_chars > 0 and len(content) > self.max_file_chars:
+                content = content[:self.max_file_chars] + "\n... (truncated)"
            
            # Try to extract text from HTML
            if '<html' in content.lower():
@@ -0,0 +1,329 @@
+"""Unified web search tool with automatic fallback across multiple providers.
+
+This module provides a single `search_web` function that automatically tries
+multiple search providers in order, falling back to the next if one fails.
+
+Search Provider Priority:
+1. Tavily (requires TAVILY_API_KEY + tavily-python)
+2. Exa (requires EXA_API_KEY + exa_py)
+3. You.com (requires YDC_API_KEY + youdotcom)
+4. DuckDuckGo (requires duckduckgo_search package, no API key)
+5. SearxNG (requires requests + running SearxNG instance)
+
+Usage:
+    from praisonaiagents.tools import search_web
+    
+    # Simple search - automatically uses best available provider
+    results = search_web("AI news 2024")
+    
+    # With max results
+    results = search_web("Python tutorials", max_results=10)
+"""
+
+from typing import List, Dict, Any, Optional
+import logging
+import os
+from importlib import util
+
+logger = logging.getLogger(__name__)
+
+
+def _check_tavily() -> tuple[bool, Optional[str]]:
+    """Check if Tavily is available."""
+    if util.find_spec("tavily") is None:
+        return False, "tavily-python package not installed"
+    if not os.environ.get("TAVILY_API_KEY"):
+        return False, "TAVILY_API_KEY not set"
+    return True, None
+
+
+def _check_exa() -> tuple[bool, Optional[str]]:
+    """Check if Exa is available."""
+    if util.find_spec("exa_py") is None:
+        return False, "exa_py package not installed"
+    if not os.environ.get("EXA_API_KEY"):
+        return False, "EXA_API_KEY not set"
+    return True, None
+
+
+def _check_youdotcom() -> tuple[bool, Optional[str]]:
+    """Check if You.com is available."""
+    if util.find_spec("youdotcom") is None:
+        return False, "youdotcom package not installed"
+    if not os.environ.get("YDC_API_KEY"):
+        return False, "YDC_API_KEY not set"
+    return True, None
+
+
+def _check_duckduckgo() -> tuple[bool, Optional[str]]:
+    """Check if DuckDuckGo is available."""
+    if util.find_spec("duckduckgo_search") is None:
+        return False, "duckduckgo_search package not installed"
+    return True, None
+
+
+def _check_searxng() -> tuple[bool, Optional[str]]:
+    """Check if SearxNG is available."""
+    if util.find_spec("requests") is None:
+        return False, "requests package not installed"
+    return True, None
+
+
+def _search_tavily(query: str, max_results: int = 5) -> List[Dict[str, Any]]:
+    """Search using Tavily."""
+    from tavily import TavilyClient
+    
+    client = TavilyClient(api_key=os.environ.get("TAVILY_API_KEY"))
+    response = client.search(query=query, max_results=max_results)
+    
+    results = []
+    for r in response.get("results", []):
+        results.append({
+            "title": r.get("title", ""),
+            "url": r.get("url", ""),
+            "snippet": r.get("content", ""),
+            "provider": "tavily"
+        })
+    return results
+
+
+def _search_exa(query: str, max_results: int = 5) -> List[Dict[str, Any]]:
+    """Search using Exa."""
+    from exa_py import Exa
+    
+    client = Exa(os.environ.get("EXA_API_KEY"))
+    response = client.search(query=query, num_results=max_results)
+    
+    results = []
+    for r in response.results:
+        results.append({
+            "title": getattr(r, "title", "") or "",
+            "url": r.url,
+            "snippet": getattr(r, "text", "") or "",
+            "provider": "exa"
+        })
+    return results
+
+
+def _search_youdotcom(query: str, max_results: int = 5) -> List[Dict[str, Any]]:
+    """Search using You.com."""
+    from youdotcom import You
+    
+    client = You(api_key_auth=os.environ.get("YDC_API_KEY"))
+    response = client.search.unified(query=query, count=max_results)
+    
+    results = []
+    # Handle response object
+    if hasattr(response, 'results'):
+        web_results = getattr(response.results, 'web', []) or []
+        for r in web_results[:max_results]:
+            results.append({
+                "title": getattr(r, "title", "") or "",
+                "url": getattr(r, "url", "") or "",
+                "snippet": getattr(r, "description", "") or "",
+                "provider": "youdotcom"
+            })
+    elif isinstance(response, dict):
+        web_results = response.get("results", {}).get("web", [])
+        for r in web_results[:max_results]:
+            results.append({
+                "title": r.get("title", ""),
+                "url": r.get("url", ""),
+                "snippet": r.get("description", ""),
+                "provider": "youdotcom"
+            })
+    return results
+
+
+def _search_duckduckgo(query: str, max_results: int = 5) -> List[Dict[str, Any]]:
+    """Search using DuckDuckGo."""
+    from duckduckgo_search import DDGS
+    
+    results = []
+    ddgs = DDGS()
+    for result in ddgs.text(keywords=query, max_results=max_results):
+        results.append({
+            "title": result.get("title", ""),
+            "url": result.get("href", ""),
+            "snippet": result.get("body", ""),
+            "provider": "duckduckgo"
+        })
+    return results
+
+
+def _search_searxng(query: str, max_results: int = 5, searxng_url: Optional[str] = None) -> List[Dict[str, Any]]:
+    """Search using SearxNG."""
+    import requests
+    
+    url = searxng_url or os.environ.get("SEARXNG_URL", "http://localhost:32768/search")
+    
+    params = {
+        'q': query,
+        'format': 'json',
+        'engines': 'google,bing,duckduckgo',
+        'safesearch': '1'
+    }
+    
+    response = requests.get(url, params=params, timeout=10)
+    response.raise_for_status()
+    
+    raw_results = response.json().get('results', [])
+    
+    results = []
+    for result in raw_results[:max_results]:
+        results.append({
+            "title": result.get("title", ""),
+            "url": result.get("url", ""),
+            "snippet": result.get("content", ""),
+            "provider": "searxng"
+        })
+    return results
+
+
+# Provider configuration: (name, check_func, search_func)
+SEARCH_PROVIDERS = [
+    ("tavily", _check_tavily, _search_tavily),
+    ("exa", _check_exa, _search_exa),
+    ("youdotcom", _check_youdotcom, _search_youdotcom),
+    ("duckduckgo", _check_duckduckgo, _search_duckduckgo),
+    ("searxng", _check_searxng, _search_searxng),
+]
+
+
+def search_web(
+    query: str,
+    max_results: int = 5,
+    providers: Optional[List[str]] = None,
+    searxng_url: Optional[str] = None,
+) -> List[Dict[str, Any]]:
+    """Search the web using multiple providers with automatic fallback.
+    
+    Tries each search provider in order until one succeeds. For each provider,
+    it first checks if the API key is set (if required) and if the package is
+    installed, then attempts the search. If the search fails, it moves to the
+    next provider.
+    
+    Args:
+        query: The search query string
+        max_results: Maximum number of results to return (default: 5)
+        providers: Optional list of provider names to try, in order.
+                  If not specified, uses default order:
+                  ["tavily", "exa", "youdotcom", "duckduckgo", "searxng"]
+        searxng_url: Optional custom SearxNG instance URL
+        
+    Returns:
+        List of search results, each containing:
+        - title: Result title
+        - url: Result URL
+        - snippet: Result description/snippet
+        - provider: Name of the provider that returned the result
+        
+        If all providers fail, returns a list with a single error dict.
+        
+    Example:
+        # Use default provider order
+        results = search_web("AI news 2024")
+        
+        # Specify providers to try
+        results = search_web("Python tutorials", providers=["duckduckgo", "tavily"])
+        
+        # With custom SearxNG URL
+        results = search_web("tech news", searxng_url="http://my-searxng:8080/search")
+    """
+    errors = []
+    
+    # Determine which providers to try
+    if providers:
+        # Filter to only valid provider names
+        provider_order = []
+        for name in providers:
+            for pname, check_fn, search_fn in SEARCH_PROVIDERS:
+                if pname == name.lower():
+                    provider_order.append((pname, check_fn, search_fn))
+                    break
+    else:
+        provider_order = SEARCH_PROVIDERS
+    
+    for provider_name, check_func, search_func in provider_order:
+        # Step 1: Check if provider is available (API key + package)
+        is_available, check_error = check_func()
+        if not is_available:
+            logger.debug(f"Skipping {provider_name}: {check_error}")
+            errors.append(f"{provider_name}: {check_error}")
+            continue
+        
+        # Step 2: Try to search
+        try:
+            logger.debug(f"Trying {provider_name} for query: {query}")
+            
+            if provider_name == "searxng":
+                results = search_func(query, max_results, searxng_url)
+            else:
+                results = search_func(query, max_results)
+            
+            if results:
+                logger.info(f"search_web: Successfully used {provider_name}")
+                return results
+            else:
+                errors.append(f"{provider_name}: No results returned")
+                
+        except Exception as e:
+            error_msg = f"{provider_name}: {str(e)}"
+            logger.debug(f"Search failed with {provider_name}: {e}")
+            errors.append(error_msg)
+            continue
+    
+    # All providers failed
+    error_summary = "; ".join(errors) if errors else "No search providers available"
+    logger.error(f"search_web: All providers failed - {error_summary}")
+    return [{"error": f"All search providers failed: {error_summary}"}]
+
+
+def get_available_providers() -> List[Dict[str, Any]]:
+    """Get list of available search providers and their status.
+    
+    Returns:
+        List of dicts with provider info:
+        - name: Provider name
+        - available: Whether the provider is ready to use
+        - reason: Reason if not available
+        
+    Example:
+        providers = get_available_providers()
+        for p in providers:
+            print(f"{p['name']}: {'✓' if p['available'] else '✗'} {p.get('reason', '')}")
+    """
+    result = []
+    for name, check_func, _ in SEARCH_PROVIDERS:
+        is_available, error = check_func()
+        result.append({
+            "name": name,
+            "available": is_available,
+            "reason": error if not is_available else None
+        })
+    return result
+
+
+if __name__ == "__main__":
+    # Example usage and provider status check
+    print("=" * 60)
+    print("Web Search Tool - Provider Status")
+    print("=" * 60)
+    
+    providers = get_available_providers()
+    for p in providers:
+        status = "✓ Available" if p["available"] else f"✗ {p['reason']}"
+        print(f"  {p['name']:12} {status}")
+    
+    print("\n" + "=" * 60)
+    print("Testing search_web...")
+    print("=" * 60)
+    
+    results = search_web("Python programming", max_results=3)
+    if results and "error" not in results[0]:
+        print(f"\nFound {len(results)} results using {results[0].get('provider', 'unknown')}:")
+        for r in results:
+            print(f"\n  Title: {r.get('title', 'N/A')[:50]}...")
+            print(f"  URL: {r.get('url', 'N/A')}")
+    else:
+        print(f"\nSearch failed: {results}")
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"

 [project]
 name = "praisonaiagents"
-version = "0.1.17"
+version = "0.1.23"
 description = "Praison AI agents for completing complex tasks with Self Reflection Agents"
 readme = "README.md"
 requires-python = ">=3.10"
@@ -0,0 +1,217 @@
+"""
+Unit tests for configurable max_file_chars in MentionsParser.
+
+TDD approach: Write tests first, then implement.
+"""
+
+import os
+import pytest
+import tempfile
+from pathlib import Path
+from unittest.mock import patch
+
+
+class TestMentionsParserConfigurableLimit:
+    """Test suite for configurable file truncation limit."""
+    
+    @pytest.fixture
+    def temp_workspace(self):
+        """Create a temporary workspace with test files."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            yield Path(tmpdir)
+    
+    @pytest.fixture
+    def small_file(self, temp_workspace):
+        """Create a small test file (1KB)."""
+        file_path = temp_workspace / "small.py"
+        content = "# Small file\n" + "x = 1\n" * 100  # ~700 chars
+        file_path.write_text(content)
+        return file_path
+    
+    @pytest.fixture
+    def large_file(self, temp_workspace):
+        """Create a large test file (100KB)."""
+        file_path = temp_workspace / "large.py"
+        content = "# Large file\n" + "x = 'a' * 100\n" * 7000  # ~100KB
+        file_path.write_text(content)
+        return file_path
+    
+    @pytest.fixture
+    def huge_file(self, temp_workspace):
+        """Create a huge test file (600KB)."""
+        file_path = temp_workspace / "huge.py"
+        content = "# Huge file\n" + "x = 'a' * 100\n" * 42000  # ~600KB
+        file_path.write_text(content)
+        return file_path
+
+    # ===== Test 1.1: Default value (500K) =====
+    def test_default_max_file_chars_is_500k(self, temp_workspace):
+        """Default max_file_chars should be 500,000."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        parser = MentionsParser(workspace_path=str(temp_workspace))
+        assert parser.max_file_chars == 500000
+    
+    # ===== Test 1.2: Custom value via constructor =====
+    def test_custom_max_file_chars_via_constructor(self, temp_workspace):
+        """max_file_chars should be configurable via constructor."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        parser = MentionsParser(
+            workspace_path=str(temp_workspace),
+            max_file_chars=100000
+        )
+        assert parser.max_file_chars == 100000
+    
+    def test_custom_max_file_chars_zero_means_no_limit(self, temp_workspace):
+        """max_file_chars=0 should mean no truncation."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        parser = MentionsParser(
+            workspace_path=str(temp_workspace),
+            max_file_chars=0
+        )
+        assert parser.max_file_chars == 0
+    
+    # ===== Test 1.3: Environment variable override =====
+    def test_env_var_override(self, temp_workspace):
+        """PRAISON_MAX_FILE_CHARS env var should override default."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        with patch.dict(os.environ, {'PRAISON_MAX_FILE_CHARS': '200000'}):
+            # Need to reimport to pick up env var
+            parser = MentionsParser(workspace_path=str(temp_workspace))
+            assert parser.max_file_chars == 200000
+    
+    def test_constructor_takes_precedence_over_env_var(self, temp_workspace):
+        """Constructor value should take precedence over env var."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        with patch.dict(os.environ, {'PRAISON_MAX_FILE_CHARS': '200000'}):
+            parser = MentionsParser(
+                workspace_path=str(temp_workspace),
+                max_file_chars=100000
+            )
+            assert parser.max_file_chars == 100000
+    
+    # ===== Test 1.4: No-limit mode =====
+    def test_no_truncation_when_limit_is_zero(self, temp_workspace, huge_file):
+        """File should not be truncated when max_file_chars=0."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        parser = MentionsParser(
+            workspace_path=str(temp_workspace),
+            max_file_chars=0
+        )
+        
+        context, _ = parser.process(f"@file:{huge_file.name} explain")
+        
+        # Should contain full content, no truncation message
+        assert "(truncated)" not in context
+        # Should have the full file content
+        original_size = huge_file.read_text().__len__()
+        # Context includes header and code fence, so check it's close
+        assert len(context) >= original_size
+    
+    # ===== Test 1.5: Truncation behavior =====
+    def test_truncation_at_custom_limit(self, temp_workspace, large_file):
+        """File should be truncated at custom limit."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        parser = MentionsParser(
+            workspace_path=str(temp_workspace),
+            max_file_chars=10000  # 10KB limit
+        )
+        
+        context, _ = parser.process(f"@file:{large_file.name} explain")
+        
+        # Should be truncated
+        assert "(truncated)" in context
+        # Content should be around the limit (plus header/fence overhead)
+        assert len(context) < 15000  # Some overhead for markdown
+    
+    def test_no_truncation_when_file_under_limit(self, temp_workspace, small_file):
+        """Small files should not be truncated."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        parser = MentionsParser(
+            workspace_path=str(temp_workspace),
+            max_file_chars=500000
+        )
+        
+        context, _ = parser.process(f"@file:{small_file.name} explain")
+        
+        # Should NOT be truncated
+        assert "(truncated)" not in context
+    
+    def test_truncation_warning_logged(self, temp_workspace, large_file, caplog):
+        """Warning should be logged when file is truncated."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        import logging
+        
+        parser = MentionsParser(
+            workspace_path=str(temp_workspace),
+            max_file_chars=10000,
+            verbose=1
+        )
+        
+        with caplog.at_level(logging.WARNING):
+            parser.process(f"@file:{large_file.name} explain")
+        
+        # Should log a warning about truncation
+        assert any("truncat" in record.message.lower() for record in caplog.records)
+    
+    # ===== Test: URL truncation also uses configurable limit =====
+    def test_url_truncation_uses_same_limit(self, temp_workspace):
+        """URL content should also respect max_file_chars."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        parser = MentionsParser(
+            workspace_path=str(temp_workspace),
+            max_file_chars=100000
+        )
+        
+        # URL truncation should use the same limit
+        assert parser.max_file_chars == 100000
+    
+    # ===== Test: Backward compatibility =====
+    def test_backward_compatible_no_args(self, temp_workspace):
+        """Parser should work without max_file_chars argument."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        
+        # Should not raise any errors
+        parser = MentionsParser(workspace_path=str(temp_workspace))
+        assert hasattr(parser, 'max_file_chars')
+    
+    # ===== Test: Performance - no impact on small files =====
+    def test_performance_small_file(self, temp_workspace, small_file):
+        """Processing small files should be fast."""
+        from praisonaiagents.tools.mentions import MentionsParser
+        import time
+        
+        parser = MentionsParser(workspace_path=str(temp_workspace))
+        
+        start = time.time()
+        for _ in range(100):
+            parser.process(f"@file:{small_file.name} explain")
+        elapsed = time.time() - start
+        
+        # 100 iterations should complete in under 1 second
+        assert elapsed < 1.0, f"Performance regression: {elapsed}s for 100 iterations"
+
+
+class TestProcessMentionsFunction:
+    """Test the convenience function."""
+    
+    def test_process_mentions_uses_default_limit(self):
+        """process_mentions function should use default limit."""
+        from praisonaiagents.tools.mentions import process_mentions
+        
+        # The function should create a parser with default settings
+        # Just verify it works without errors
+        context, cleaned = process_mentions("hello world")
+        assert cleaned == "hello world"
+
+
+if __name__ == "__main__":
+    pytest.main([__file__, "-v"])
@@ -0,0 +1 @@
+"""Unit tests for PraisonAI Agents evaluation framework."""
@@ -0,0 +1,146 @@
+"""Unit tests for BaseEvaluator class."""
+
+import pytest
+import tempfile
+import json
+from pathlib import Path
+from unittest.mock import MagicMock, patch
+
+from praisonaiagents.eval.base import BaseEvaluator
+
+
+class ConcreteEvaluator(BaseEvaluator):
+    """Concrete implementation for testing."""
+    
+    def run(self, **kwargs):
+        return {"test": "result"}
+
+
+class TestBaseEvaluator:
+    """Tests for BaseEvaluator abstract class."""
+    
+    def test_init_default_values(self):
+        """Test default initialization values."""
+        evaluator = ConcreteEvaluator()
+        assert evaluator.name.startswith("eval_")
+        assert evaluator.eval_id is not None
+        assert len(evaluator.eval_id) == 8
+        assert evaluator.save_results_path is None
+        assert evaluator.verbose is False
+    
+    def test_init_custom_values(self):
+        """Test initialization with custom values."""
+        evaluator = ConcreteEvaluator(
+            name="my_eval",
+            save_results_path="/tmp/results.json",
+            verbose=True
+        )
+        assert evaluator.name == "my_eval"
+        assert evaluator.save_results_path == "/tmp/results.json"
+        assert evaluator.verbose is True
+    
+    def test_eval_id_unique(self):
+        """Test that each evaluator gets a unique ID."""
+        eval1 = ConcreteEvaluator()
+        eval2 = ConcreteEvaluator()
+        assert eval1.eval_id != eval2.eval_id
+    
+    def test_before_run_verbose(self):
+        """Test before_run hook with verbose logging."""
+        evaluator = ConcreteEvaluator(verbose=True)
+        with patch('praisonaiagents.eval.base.logger') as mock_logger:
+            evaluator.before_run()
+            mock_logger.info.assert_called()
+    
+    def test_after_run_verbose(self):
+        """Test after_run hook with verbose logging."""
+        evaluator = ConcreteEvaluator(verbose=True)
+        with patch('praisonaiagents.eval.base.logger') as mock_logger:
+            evaluator.after_run({"result": "test"})
+            mock_logger.info.assert_called()
+    
+    def test_save_result_to_file(self):
+        """Test saving results to file."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/results.json"
+            evaluator = ConcreteEvaluator(
+                name="test_save",
+                save_results_path=path
+            )
+            
+            result = MagicMock()
+            result.to_dict.return_value = {"score": 8.0}
+            
+            evaluator._save_result(result)
+            
+            assert Path(path).exists()
+            with open(path) as f:
+                data = json.load(f)
+            assert data["score"] == 8.0
+    
+    def test_save_result_with_placeholders(self):
+        """Test saving results with path placeholders."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/{{name}}_{{eval_id}}.json"
+            evaluator = ConcreteEvaluator(
+                name="test_eval",
+                save_results_path=path
+            )
+            
+            result = MagicMock()
+            result.to_dict.return_value = {"score": 8.0}
+            
+            evaluator._save_result(result)
+            
+            expected_path = f"{tmpdir}/test_eval_{evaluator.eval_id}.json"
+            assert Path(expected_path).exists()
+    
+    def test_save_result_creates_directories(self):
+        """Test that save creates parent directories."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/nested/dir/results.json"
+            evaluator = ConcreteEvaluator(save_results_path=path)
+            
+            result = MagicMock()
+            result.to_dict.return_value = {"score": 8.0}
+            
+            evaluator._save_result(result)
+            
+            assert Path(path).exists()
+    
+    def test_run_abstract_method(self):
+        """Test that run method works in concrete implementation."""
+        evaluator = ConcreteEvaluator()
+        result = evaluator.run()
+        assert result == {"test": "result"}
+    
+    @pytest.mark.asyncio
+    async def test_run_async_default(self):
+        """Test default async run calls sync run."""
+        evaluator = ConcreteEvaluator()
+        result = await evaluator.run_async()
+        assert result == {"test": "result"}
+    
+    @pytest.mark.asyncio
+    async def test_async_before_run(self):
+        """Test async before_run hook."""
+        evaluator = ConcreteEvaluator(verbose=True)
+        with patch('praisonaiagents.eval.base.logger') as mock_logger:
+            await evaluator.async_before_run()
+            mock_logger.info.assert_called()
+    
+    @pytest.mark.asyncio
+    async def test_async_after_run(self):
+        """Test async after_run hook."""
+        evaluator = ConcreteEvaluator(verbose=True)
+        with patch('praisonaiagents.eval.base.logger') as mock_logger:
+            await evaluator.async_after_run({"result": "test"})
+            mock_logger.info.assert_called()
+    
+    def test_repr(self):
+        """Test string representation."""
+        evaluator = ConcreteEvaluator(name="test_eval")
+        repr_str = repr(evaluator)
+        assert "ConcreteEvaluator" in repr_str
+        assert "test_eval" in repr_str
+        assert evaluator.eval_id in repr_str
@@ -0,0 +1,425 @@
+"""Unit tests for evaluator classes."""
+
+import pytest
+import time
+from unittest.mock import MagicMock, patch, AsyncMock
+
+from praisonaiagents.eval.accuracy import AccuracyEvaluator
+from praisonaiagents.eval.performance import PerformanceEvaluator
+from praisonaiagents.eval.reliability import ReliabilityEvaluator
+from praisonaiagents.eval.criteria import CriteriaEvaluator
+from praisonaiagents.eval.results import (
+    AccuracyResult,
+    PerformanceResult,
+    ReliabilityResult,
+    CriteriaResult,
+)
+
+
+class TestAccuracyEvaluator:
+    """Tests for AccuracyEvaluator class."""
+    
+    def test_init_with_func(self):
+        """Test initialization with function."""
+        func = lambda x: "test output"
+        evaluator = AccuracyEvaluator(
+            func=func,
+            input_text="test input",
+            expected_output="test output"
+        )
+        assert evaluator.func == func
+        assert evaluator.input_text == "test input"
+        assert evaluator.expected_output == "test output"
+    
+    def test_init_with_agent(self):
+        """Test initialization with agent."""
+        agent = MagicMock()
+        evaluator = AccuracyEvaluator(
+            agent=agent,
+            input_text="test",
+            expected_output="expected"
+        )
+        assert evaluator.agent == agent
+    
+    def test_init_requires_agent_or_func(self):
+        """Test that either agent or func must be provided."""
+        with pytest.raises(ValueError, match="Either 'agent' or 'func' must be provided"):
+            AccuracyEvaluator(
+                input_text="test",
+                expected_output="expected"
+            )
+    
+    def test_get_output_from_func(self):
+        """Test getting output from function."""
+        func = lambda x: f"output for {x}"
+        evaluator = AccuracyEvaluator(
+            func=func,
+            input_text="test",
+            expected_output="expected"
+        )
+        output = evaluator._get_output()
+        assert output == "output for test"
+    
+    def test_get_output_from_agent_chat(self):
+        """Test getting output from agent with chat method."""
+        agent = MagicMock()
+        agent.chat.return_value = "agent response"
+        
+        evaluator = AccuracyEvaluator(
+            agent=agent,
+            input_text="test",
+            expected_output="expected"
+        )
+        output = evaluator._get_output()
+        assert output == "agent response"
+        agent.chat.assert_called_once_with("test")
+    
+    def test_get_output_from_agent_start(self):
+        """Test getting output from agent with start method."""
+        agent = MagicMock(spec=['start'])
+        result = MagicMock()
+        result.raw = "agent result"
+        agent.start.return_value = result
+        
+        evaluator = AccuracyEvaluator(
+            agent=agent,
+            input_text="test",
+            expected_output="expected"
+        )
+        output = evaluator._get_output()
+        assert output == "agent result"
+    
+    def test_judge_output(self):
+        """Test LLM judging of output."""
+        with patch('openai.OpenAI') as mock_openai_class:
+            mock_client = MagicMock()
+            mock_openai_class.return_value = mock_client
+            
+            mock_response = MagicMock()
+            mock_response.choices = [MagicMock()]
+            mock_response.choices[0].message.content = "SCORE: 8\nREASONING: Good match"
+            mock_client.chat.completions.create.return_value = mock_response
+            
+            def test_func(x):
+                return "4"
+            
+            evaluator = AccuracyEvaluator(
+                func=test_func,
+                input_text="What is 2+2?",
+                expected_output="4"
+            )
+            
+            score = evaluator._judge_output("4")
+            assert score.score == 8.0
+            assert "Good match" in score.reasoning
+    
+    def test_run_returns_accuracy_result(self):
+        """Test that run returns AccuracyResult."""
+        with patch('openai.OpenAI') as mock_openai_class:
+            mock_client = MagicMock()
+            mock_openai_class.return_value = mock_client
+            
+            mock_response = MagicMock()
+            mock_response.choices = [MagicMock()]
+            mock_response.choices[0].message.content = "SCORE: 9\nREASONING: Excellent"
+            mock_client.chat.completions.create.return_value = mock_response
+            
+            def test_func(x):
+                return "4"
+            
+            evaluator = AccuracyEvaluator(
+                func=test_func,
+                input_text="What is 2+2?",
+                expected_output="4",
+                num_iterations=2
+            )
+            
+            result = evaluator.run()
+            assert isinstance(result, AccuracyResult)
+            assert len(result.evaluations) == 2
+    
+    def test_evaluate_output(self):
+        """Test evaluate_output method for pre-generated outputs."""
+        with patch('openai.OpenAI') as mock_openai_class:
+            mock_client = MagicMock()
+            mock_openai_class.return_value = mock_client
+            
+            mock_response = MagicMock()
+            mock_response.choices = [MagicMock()]
+            mock_response.choices[0].message.content = "SCORE: 10\nREASONING: Perfect"
+            mock_client.chat.completions.create.return_value = mock_response
+            
+            def test_func(x):
+                return "unused"
+            
+            evaluator = AccuracyEvaluator(
+                func=test_func,
+                input_text="What is 2+2?",
+                expected_output="4"
+            )
+            
+            result = evaluator.evaluate_output("4")
+            assert isinstance(result, AccuracyResult)
+            assert len(result.evaluations) == 1
+            assert result.evaluations[0].score == 10.0
+
+
+class TestPerformanceEvaluator:
+    """Tests for PerformanceEvaluator class."""
+    
+    def test_init_with_func(self):
+        """Test initialization with function."""
+        func = lambda: time.sleep(0.01)
+        evaluator = PerformanceEvaluator(
+            func=func,
+            num_iterations=5,
+            warmup_runs=2
+        )
+        assert evaluator.func == func
+        assert evaluator.num_iterations == 5
+        assert evaluator.warmup_runs == 2
+    
+    def test_init_with_agent(self):
+        """Test initialization with agent."""
+        agent = MagicMock()
+        evaluator = PerformanceEvaluator(
+            agent=agent,
+            input_text="test",
+            num_iterations=3
+        )
+        assert evaluator.agent == agent
+        assert evaluator.input_text == "test"
+    
+    def test_init_requires_func_or_agent(self):
+        """Test that either func or agent must be provided."""
+        with pytest.raises(ValueError, match="Either 'func' or 'agent' must be provided"):
+            PerformanceEvaluator()
+    
+    def test_run_measures_time(self):
+        """Test that run measures execution time."""
+        call_count = 0
+        def slow_func():
+            nonlocal call_count
+            call_count += 1
+            time.sleep(0.01)
+        
+        evaluator = PerformanceEvaluator(
+            func=slow_func,
+            num_iterations=3,
+            warmup_runs=1,
+            track_memory=False
+        )
+        
+        result = evaluator.run()
+        assert isinstance(result, PerformanceResult)
+        assert len(result.metrics) == 3
+        assert all(m.run_time >= 0.01 for m in result.metrics)
+        assert call_count == 4  # 1 warmup + 3 iterations
+    
+    def test_run_with_agent(self):
+        """Test run with agent."""
+        agent = MagicMock()
+        agent.chat.return_value = "response"
+        
+        evaluator = PerformanceEvaluator(
+            agent=agent,
+            input_text="test",
+            num_iterations=2,
+            warmup_runs=1,
+            track_memory=False
+        )
+        
+        result = evaluator.run()
+        assert isinstance(result, PerformanceResult)
+        assert len(result.metrics) == 2
+        assert agent.chat.call_count == 3  # 1 warmup + 2 iterations
+
+
+class TestReliabilityEvaluator:
+    """Tests for ReliabilityEvaluator class."""
+    
+    def test_init(self):
+        """Test initialization."""
+        agent = MagicMock()
+        evaluator = ReliabilityEvaluator(
+            agent=agent,
+            input_text="test",
+            expected_tools=["search", "calculate"]
+        )
+        assert evaluator.agent == agent
+        assert evaluator.expected_tools == {"search", "calculate"}
+    
+    def test_init_requires_agent(self):
+        """Test that agent must be provided."""
+        with pytest.raises(ValueError, match="'agent' must be provided"):
+            ReliabilityEvaluator(
+                input_text="test",
+                expected_tools=["search"]
+            )
+    
+    def test_evaluate_tool_calls(self):
+        """Test evaluate_tool_calls method."""
+        agent = MagicMock()
+        evaluator = ReliabilityEvaluator(
+            agent=agent,
+            input_text="test",
+            expected_tools=["search", "calculate"],
+            forbidden_tools=["delete"]
+        )
+        
+        result = evaluator.evaluate_tool_calls(["search", "calculate"])
+        assert isinstance(result, ReliabilityResult)
+        assert result.passed is True
+        assert len(result.passed_calls) == 3  # 2 expected + 1 forbidden not called
+    
+    def test_evaluate_tool_calls_with_failures(self):
+        """Test evaluate_tool_calls with missing tools."""
+        agent = MagicMock()
+        evaluator = ReliabilityEvaluator(
+            agent=agent,
+            input_text="test",
+            expected_tools=["search", "calculate"]
+        )
+        
+        result = evaluator.evaluate_tool_calls(["search"])  # missing calculate
+        assert result.passed is False
+        assert len(result.failed_calls) == 1
+    
+    def test_forbidden_tools(self):
+        """Test forbidden tools detection."""
+        agent = MagicMock()
+        evaluator = ReliabilityEvaluator(
+            agent=agent,
+            input_text="test",
+            expected_tools=[],
+            forbidden_tools=["delete", "drop"]
+        )
+        
+        result = evaluator.evaluate_tool_calls(["delete"])  # called forbidden tool
+        assert result.passed is False
+        assert len(result.failed_calls) == 1
+
+
+class TestCriteriaEvaluator:
+    """Tests for CriteriaEvaluator class."""
+    
+    def test_init(self):
+        """Test initialization."""
+        evaluator = CriteriaEvaluator(
+            criteria="Response is helpful",
+            func=lambda x: "test",
+            input_text="test"
+        )
+        assert evaluator.criteria == "Response is helpful"
+        assert evaluator.scoring_type == "numeric"
+        assert evaluator.threshold == 7.0
+    
+    def test_init_binary_scoring(self):
+        """Test initialization with binary scoring."""
+        evaluator = CriteriaEvaluator(
+            criteria="Response is helpful",
+            func=lambda x: "test",
+            scoring_type="binary"
+        )
+        assert evaluator.scoring_type == "binary"
+    
+    def test_judge_output_numeric(self):
+        """Test numeric scoring."""
+        with patch('openai.OpenAI') as mock_openai_class:
+            mock_client = MagicMock()
+            mock_openai_class.return_value = mock_client
+            
+            mock_response = MagicMock()
+            mock_response.choices = [MagicMock()]
+            mock_response.choices[0].message.content = "SCORE: 8\nREASONING: Good"
+            mock_client.chat.completions.create.return_value = mock_response
+            
+            def test_func(x):
+                return "test"
+            
+            evaluator = CriteriaEvaluator(
+                criteria="Be helpful",
+                func=test_func,
+                scoring_type="numeric",
+                threshold=7.0
+            )
+            
+            score = evaluator._judge_output("test output")
+            assert score.score == 8.0
+            assert score.passed is True
+    
+    def test_judge_output_binary(self):
+        """Test binary scoring."""
+        with patch('openai.OpenAI') as mock_openai_class:
+            mock_client = MagicMock()
+            mock_openai_class.return_value = mock_client
+            
+            mock_response = MagicMock()
+            mock_response.choices = [MagicMock()]
+            mock_response.choices[0].message.content = "RESULT: PASS\nREASONING: Meets criteria"
+            mock_client.chat.completions.create.return_value = mock_response
+            
+            def test_func(x):
+                return "test"
+            
+            evaluator = CriteriaEvaluator(
+                criteria="Be helpful",
+                func=test_func,
+                scoring_type="binary"
+            )
+            
+            score = evaluator._judge_output("test output")
+            assert score.score == 10.0
+            assert score.passed is True
+    
+    def test_on_fail_callback(self):
+        """Test on_fail callback is called."""
+        with patch('openai.OpenAI') as mock_openai_class:
+            mock_client = MagicMock()
+            mock_openai_class.return_value = mock_client
+            
+            mock_response = MagicMock()
+            mock_response.choices = [MagicMock()]
+            mock_response.choices[0].message.content = "SCORE: 3\nREASONING: Poor"
+            mock_client.chat.completions.create.return_value = mock_response
+            
+            on_fail_called = []
+            def on_fail(score):
+                on_fail_called.append(score)
+            
+            def test_func(x):
+                return "test"
+            
+            evaluator = CriteriaEvaluator(
+                criteria="Be helpful",
+                func=test_func,
+                threshold=7.0,
+                on_fail=on_fail
+            )
+            
+            evaluator.run()
+            assert len(on_fail_called) == 1
+            assert on_fail_called[0].score == 3.0
+    
+    def test_evaluate_output_criteria(self):
+        """Test evaluate_output method."""
+        with patch('openai.OpenAI') as mock_openai_class:
+            mock_client = MagicMock()
+            mock_openai_class.return_value = mock_client
+            
+            mock_response = MagicMock()
+            mock_response.choices = [MagicMock()]
+            mock_response.choices[0].message.content = "SCORE: 9\nREASONING: Great"
+            mock_client.chat.completions.create.return_value = mock_response
+            
+            def test_func(x):
+                return "unused"
+            
+            evaluator = CriteriaEvaluator(
+                criteria="Be helpful",
+                func=test_func
+            )
+            
+            result = evaluator.evaluate_output("pre-generated output")
+            assert isinstance(result, CriteriaResult)
+            assert len(result.evaluations) == 1
@@ -0,0 +1,359 @@
+"""Unit tests for evaluation result dataclasses."""
+
+import pytest
+import json
+from datetime import datetime
+
+from praisonaiagents.eval.results import (
+    EvaluationScore,
+    AccuracyResult,
+    PerformanceMetrics,
+    PerformanceResult,
+    ToolCallResult,
+    ReliabilityResult,
+    CriteriaScore,
+    CriteriaResult,
+)
+
+
+class TestEvaluationScore:
+    """Tests for EvaluationScore dataclass."""
+    
+    def test_creation(self):
+        """Test basic creation of EvaluationScore."""
+        score = EvaluationScore(
+            score=8.5,
+            reasoning="Good response",
+            input_text="What is 2+2?",
+            output_text="4",
+            expected_output="4"
+        )
+        assert score.score == 8.5
+        assert score.reasoning == "Good response"
+        assert score.input_text == "What is 2+2?"
+        assert score.output_text == "4"
+        assert score.expected_output == "4"
+    
+    def test_timestamp_auto_generated(self):
+        """Test that timestamp is auto-generated."""
+        score = EvaluationScore(
+            score=8.0,
+            reasoning="Test",
+            input_text="test",
+            output_text="test"
+        )
+        assert score.timestamp is not None
+        datetime.fromisoformat(score.timestamp)
+    
+    def test_metadata_default(self):
+        """Test that metadata defaults to empty dict."""
+        score = EvaluationScore(
+            score=8.0,
+            reasoning="Test",
+            input_text="test",
+            output_text="test"
+        )
+        assert score.metadata == {}
+
+
+class TestAccuracyResult:
+    """Tests for AccuracyResult dataclass."""
+    
+    def test_empty_result(self):
+        """Test empty AccuracyResult."""
+        result = AccuracyResult()
+        assert result.evaluations == []
+        assert result.avg_score == 0.0
+        assert result.min_score == 0.0
+        assert result.max_score == 0.0
+        assert result.std_dev == 0.0
+        assert result.passed is False
+    
+    def test_single_evaluation(self):
+        """Test AccuracyResult with single evaluation."""
+        result = AccuracyResult()
+        result.evaluations.append(EvaluationScore(
+            score=8.0,
+            reasoning="Good",
+            input_text="test",
+            output_text="test"
+        ))
+        assert result.avg_score == 8.0
+        assert result.min_score == 8.0
+        assert result.max_score == 8.0
+        assert result.std_dev == 0.0
+        assert result.passed is True
+    
+    def test_multiple_evaluations(self):
+        """Test AccuracyResult with multiple evaluations."""
+        result = AccuracyResult()
+        for score in [6.0, 8.0, 10.0]:
+            result.evaluations.append(EvaluationScore(
+                score=score,
+                reasoning="Test",
+                input_text="test",
+                output_text="test"
+            ))
+        assert result.avg_score == 8.0
+        assert result.min_score == 6.0
+        assert result.max_score == 10.0
+        assert result.passed is True
+    
+    def test_passed_threshold(self):
+        """Test passed property with different scores."""
+        result = AccuracyResult()
+        result.evaluations.append(EvaluationScore(
+            score=6.0,
+            reasoning="Below threshold",
+            input_text="test",
+            output_text="test"
+        ))
+        assert result.passed is False
+        
+        result.evaluations.append(EvaluationScore(
+            score=8.0,
+            reasoning="Above threshold",
+            input_text="test",
+            output_text="test"
+        ))
+        assert result.passed is True
+    
+    def test_to_dict(self):
+        """Test to_dict method."""
+        result = AccuracyResult(eval_id="test123", name="test_eval")
+        result.evaluations.append(EvaluationScore(
+            score=8.0,
+            reasoning="Good",
+            input_text="test",
+            output_text="test"
+        ))
+        d = result.to_dict()
+        assert d["eval_id"] == "test123"
+        assert d["name"] == "test_eval"
+        assert d["avg_score"] == 8.0
+        assert d["num_evaluations"] == 1
+    
+    def test_to_json(self):
+        """Test to_json method."""
+        result = AccuracyResult(eval_id="test123")
+        result.evaluations.append(EvaluationScore(
+            score=8.0,
+            reasoning="Good",
+            input_text="test",
+            output_text="test"
+        ))
+        json_str = result.to_json()
+        data = json.loads(json_str)
+        assert data["eval_id"] == "test123"
+
+
+class TestPerformanceMetrics:
+    """Tests for PerformanceMetrics dataclass."""
+    
+    def test_creation(self):
+        """Test basic creation."""
+        metrics = PerformanceMetrics(
+            run_time=0.5,
+            memory_usage=10.5,
+            iteration=1
+        )
+        assert metrics.run_time == 0.5
+        assert metrics.memory_usage == 10.5
+        assert metrics.iteration == 1
+
+
+class TestPerformanceResult:
+    """Tests for PerformanceResult dataclass."""
+    
+    def test_empty_result(self):
+        """Test empty PerformanceResult."""
+        result = PerformanceResult()
+        assert result.metrics == []
+        assert result.avg_run_time == 0.0
+        assert result.min_run_time == 0.0
+        assert result.max_run_time == 0.0
+        assert result.avg_memory == 0.0
+    
+    def test_with_metrics(self):
+        """Test PerformanceResult with metrics."""
+        result = PerformanceResult()
+        for i, (time, mem) in enumerate([(0.1, 5.0), (0.2, 10.0), (0.3, 15.0)]):
+            result.metrics.append(PerformanceMetrics(
+                run_time=time,
+                memory_usage=mem,
+                iteration=i + 1
+            ))
+        assert result.avg_run_time == 0.2
+        assert result.min_run_time == 0.1
+        assert result.max_run_time == 0.3
+        assert result.avg_memory == 10.0
+        assert result.max_memory == 15.0
+    
+    def test_median_and_p95(self):
+        """Test median and p95 calculations."""
+        result = PerformanceResult()
+        for i in range(100):
+            result.metrics.append(PerformanceMetrics(
+                run_time=float(i + 1) / 100,
+                memory_usage=1.0,
+                iteration=i + 1
+            ))
+        assert 0.49 <= result.median_run_time <= 0.51
+        assert result.p95_run_time >= 0.95
+    
+    def test_to_dict(self):
+        """Test to_dict method."""
+        result = PerformanceResult(warmup_runs=5, eval_id="perf123")
+        result.metrics.append(PerformanceMetrics(
+            run_time=0.5,
+            memory_usage=10.0,
+            iteration=1
+        ))
+        d = result.to_dict()
+        assert d["warmup_runs"] == 5
+        assert d["eval_id"] == "perf123"
+        assert d["num_iterations"] == 1
+
+
+class TestToolCallResult:
+    """Tests for ToolCallResult dataclass."""
+    
+    def test_creation(self):
+        """Test basic creation."""
+        tc = ToolCallResult(
+            tool_name="search",
+            expected=True,
+            actual=True,
+            passed=True
+        )
+        assert tc.tool_name == "search"
+        assert tc.expected is True
+        assert tc.actual is True
+        assert tc.passed is True
+
+
+class TestReliabilityResult:
+    """Tests for ReliabilityResult dataclass."""
+    
+    def test_empty_result(self):
+        """Test empty ReliabilityResult."""
+        result = ReliabilityResult()
+        assert result.tool_results == []
+        assert result.pass_rate == 0.0
+        assert result.status == "PASSED"
+        assert result.passed is True
+    
+    def test_with_tool_calls(self):
+        """Test ReliabilityResult with tool calls."""
+        result = ReliabilityResult()
+        result.tool_results.append(ToolCallResult(
+            tool_name="search",
+            expected=True,
+            actual=True,
+            passed=True
+        ))
+        result.tool_results.append(ToolCallResult(
+            tool_name="calculate",
+            expected=True,
+            actual=False,
+            passed=False
+        ))
+        assert len(result.passed_calls) == 1
+        assert len(result.failed_calls) == 1
+        assert result.pass_rate == 0.5
+        assert result.status == "FAILED"
+        assert result.passed is False
+    
+    def test_all_passed(self):
+        """Test when all tool calls pass."""
+        result = ReliabilityResult()
+        for name in ["tool1", "tool2", "tool3"]:
+            result.tool_results.append(ToolCallResult(
+                tool_name=name,
+                expected=True,
+                actual=True,
+                passed=True
+            ))
+        assert result.pass_rate == 1.0
+        assert result.passed is True
+
+
+class TestCriteriaScore:
+    """Tests for CriteriaScore dataclass."""
+    
+    def test_creation(self):
+        """Test basic creation."""
+        score = CriteriaScore(
+            score=8.0,
+            passed=True,
+            reasoning="Meets criteria",
+            output_text="test output",
+            criteria="Be helpful"
+        )
+        assert score.score == 8.0
+        assert score.passed is True
+        assert score.reasoning == "Meets criteria"
+
+
+class TestCriteriaResult:
+    """Tests for CriteriaResult dataclass."""
+    
+    def test_empty_result(self):
+        """Test empty CriteriaResult."""
+        result = CriteriaResult(criteria="Be helpful")
+        assert result.evaluations == []
+        assert result.avg_score == 0.0
+        assert result.pass_rate == 0.0
+    
+    def test_numeric_scoring(self):
+        """Test numeric scoring mode."""
+        result = CriteriaResult(
+            criteria="Be helpful",
+            scoring_type="numeric",
+            threshold=7.0
+        )
+        result.evaluations.append(CriteriaScore(
+            score=8.0,
+            passed=True,
+            reasoning="Good",
+            output_text="test",
+            criteria="Be helpful"
+        ))
+        assert result.avg_score == 8.0
+        assert result.passed is True
+    
+    def test_binary_scoring(self):
+        """Test binary scoring mode."""
+        result = CriteriaResult(
+            criteria="Be helpful",
+            scoring_type="binary"
+        )
+        result.evaluations.append(CriteriaScore(
+            score=10.0,
+            passed=True,
+            reasoning="Pass",
+            output_text="test",
+            criteria="Be helpful"
+        ))
+        result.evaluations.append(CriteriaScore(
+            score=0.0,
+            passed=False,
+            reasoning="Fail",
+            output_text="test",
+            criteria="Be helpful"
+        ))
+        assert result.pass_rate == 0.5
+        assert result.passed is True
+    
+    def test_to_dict(self):
+        """Test to_dict method."""
+        result = CriteriaResult(
+            criteria="Be helpful",
+            scoring_type="numeric",
+            threshold=7.0,
+            eval_id="crit123"
+        )
+        d = result.to_dict()
+        assert d["criteria"] == "Be helpful"
+        assert d["scoring_type"] == "numeric"
+        assert d["threshold"] == 7.0
@@ -0,0 +1,222 @@
+"""Unit tests for evaluation utility functions."""
+
+import json
+import tempfile
+from pathlib import Path
+from unittest.mock import MagicMock
+
+from praisonaiagents.eval.utils import (
+    save_result_to_file,
+    load_result_from_file,
+    format_score,
+    format_percentage,
+    format_duration,
+    format_memory,
+    calculate_pass_rate,
+)
+
+
+class TestSaveResultToFile:
+    """Tests for save_result_to_file function."""
+    
+    def test_save_with_to_dict(self):
+        """Test saving result with to_dict method."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/result.json"
+            result = MagicMock()
+            result.to_dict.return_value = {"score": 8.0, "passed": True}
+            
+            success = save_result_to_file(result, path)
+            
+            assert success is True
+            assert Path(path).exists()
+            with open(path) as f:
+                data = json.load(f)
+            assert data["score"] == 8.0
+    
+    def test_save_with_dict_attr(self):
+        """Test saving result with __dict__ attribute."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/result.json"
+            
+            class SimpleResult:
+                def __init__(self):
+                    self.score = 9.0
+                    self.name = "test"
+            
+            result = SimpleResult()
+            success = save_result_to_file(result, path)
+            
+            assert success is True
+            with open(path) as f:
+                data = json.load(f)
+            assert data["score"] == 9.0
+    
+    def test_save_with_placeholders(self):
+        """Test saving with path placeholders."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/{{name}}_{{eval_id}}.json"
+            result = MagicMock()
+            result.to_dict.return_value = {"score": 8.0}
+            
+            success = save_result_to_file(
+                result, path,
+                eval_id="abc123",
+                name="test_eval"
+            )
+            
+            assert success is True
+            expected_path = f"{tmpdir}/test_eval_abc123.json"
+            assert Path(expected_path).exists()
+    
+    def test_save_creates_directories(self):
+        """Test that save creates parent directories."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/nested/dir/result.json"
+            result = MagicMock()
+            result.to_dict.return_value = {"score": 8.0}
+            
+            success = save_result_to_file(result, path)
+            
+            assert success is True
+            assert Path(path).exists()
+    
+    def test_save_failure_returns_false(self):
+        """Test that save failure returns False."""
+        result = MagicMock()
+        result.to_dict.side_effect = Exception("Test error")
+        
+        success = save_result_to_file(result, "/invalid/path/result.json")
+        
+        assert success is False
+
+
+class TestLoadResultFromFile:
+    """Tests for load_result_from_file function."""
+    
+    def test_load_existing_file(self):
+        """Test loading existing result file."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/result.json"
+            data = {"score": 8.0, "passed": True}
+            with open(path, 'w') as f:
+                json.dump(data, f)
+            
+            loaded = load_result_from_file(path)
+            
+            assert loaded == data
+    
+    def test_load_nonexistent_file(self):
+        """Test loading nonexistent file returns None."""
+        result = load_result_from_file("/nonexistent/path/result.json")
+        assert result is None
+    
+    def test_load_invalid_json(self):
+        """Test loading invalid JSON returns None."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            path = f"{tmpdir}/invalid.json"
+            with open(path, 'w') as f:
+                f.write("not valid json")
+            
+            result = load_result_from_file(path)
+            assert result is None
+
+
+class TestFormatScore:
+    """Tests for format_score function."""
+    
+    def test_format_default_max(self):
+        """Test formatting with default max score."""
+        assert format_score(8.5) == "8.5/10"
+    
+    def test_format_custom_max(self):
+        """Test formatting with custom max score."""
+        assert format_score(85.0, 100.0) == "85.0/100"
+    
+    def test_format_zero(self):
+        """Test formatting zero score."""
+        assert format_score(0.0) == "0.0/10"
+
+
+class TestFormatPercentage:
+    """Tests for format_percentage function."""
+    
+    def test_format_full(self):
+        """Test formatting 100%."""
+        assert format_percentage(1.0) == "100.0%"
+    
+    def test_format_half(self):
+        """Test formatting 50%."""
+        assert format_percentage(0.5) == "50.0%"
+    
+    def test_format_zero(self):
+        """Test formatting 0%."""
+        assert format_percentage(0.0) == "0.0%"
+    
+    def test_format_decimal(self):
+        """Test formatting decimal percentage."""
+        assert format_percentage(0.333) == "33.3%"
+
+
+class TestFormatDuration:
+    """Tests for format_duration function."""
+    
+    def test_format_microseconds(self):
+        """Test formatting microseconds."""
+        result = format_duration(0.0001)
+        assert "μs" in result
+    
+    def test_format_milliseconds(self):
+        """Test formatting milliseconds."""
+        result = format_duration(0.5)
+        assert "ms" in result
+    
+    def test_format_seconds(self):
+        """Test formatting seconds."""
+        result = format_duration(5.5)
+        assert "s" in result
+        assert "m" not in result
+    
+    def test_format_minutes(self):
+        """Test formatting minutes."""
+        result = format_duration(90.0)
+        assert "m" in result
+
+
+class TestFormatMemory:
+    """Tests for format_memory function."""
+    
+    def test_format_kilobytes(self):
+        """Test formatting kilobytes."""
+        result = format_memory(0.5)
+        assert "KB" in result
+    
+    def test_format_megabytes(self):
+        """Test formatting megabytes."""
+        result = format_memory(100.0)
+        assert "MB" in result
+    
+    def test_format_gigabytes(self):
+        """Test formatting gigabytes."""
+        result = format_memory(2048.0)
+        assert "GB" in result
+
+
+class TestCalculatePassRate:
+    """Tests for calculate_pass_rate function."""
+    
+    def test_all_passed(self):
+        """Test 100% pass rate."""
+        assert calculate_pass_rate(10, 10) == 1.0
+    
+    def test_none_passed(self):
+        """Test 0% pass rate."""
+        assert calculate_pass_rate(0, 10) == 0.0
+    
+    def test_half_passed(self):
+        """Test 50% pass rate."""
+        assert calculate_pass_rate(5, 10) == 0.5
+    
+    def test_zero_total(self):
+        """Test with zero total."""
+        assert calculate_pass_rate(0, 0) == 0.0
@@ -0,0 +1 @@
+"""Unit tests for Agent Skills module."""
@@ -0,0 +1,180 @@
+"""Tests for Agent Skills discovery."""
+
+from pathlib import Path
+import tempfile
+
+
+class TestGetDefaultSkillDirs:
+    """Tests for get_default_skill_dirs function."""
+    
+    def test_returns_list(self):
+        """Test that function returns a list."""
+        from praisonaiagents.skills.discovery import get_default_skill_dirs
+        
+        result = get_default_skill_dirs()
+        
+        assert isinstance(result, list)
+    
+    def test_returns_path_objects(self):
+        """Test that returned items are Path objects."""
+        from praisonaiagents.skills.discovery import get_default_skill_dirs
+        
+        result = get_default_skill_dirs()
+        
+        for item in result:
+            assert isinstance(item, Path)
+
+
+class TestDiscoverSkills:
+    """Tests for discover_skills function."""
+    
+    def test_discover_empty_directory(self):
+        """Test discovering skills in empty directory."""
+        from praisonaiagents.skills.discovery import discover_skills
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            result = discover_skills([tmpdir], include_defaults=False)
+            
+            assert result == []
+    
+    def test_discover_single_skill(self):
+        """Test discovering a single valid skill."""
+        from praisonaiagents.skills.discovery import discover_skills
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            # Create a skill directory
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill for discovery
+---
+
+# Test Skill
+""")
+            
+            result = discover_skills([tmpdir], include_defaults=False)
+            
+            assert len(result) == 1
+            assert result[0].name == "test-skill"
+            assert result[0].description == "A test skill for discovery"
+    
+    def test_discover_multiple_skills(self):
+        """Test discovering multiple skills."""
+        from praisonaiagents.skills.discovery import discover_skills
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            # Create first skill
+            skill1_dir = Path(tmpdir) / "skill-one"
+            skill1_dir.mkdir()
+            (skill1_dir / "SKILL.md").write_text("""---
+name: skill-one
+description: First skill
+---
+""")
+            
+            # Create second skill
+            skill2_dir = Path(tmpdir) / "skill-two"
+            skill2_dir.mkdir()
+            (skill2_dir / "SKILL.md").write_text("""---
+name: skill-two
+description: Second skill
+---
+""")
+            
+            result = discover_skills([tmpdir], include_defaults=False)
+            
+            assert len(result) == 2
+            names = {s.name for s in result}
+            assert "skill-one" in names
+            assert "skill-two" in names
+    
+    def test_discover_skips_invalid_skills(self):
+        """Test that invalid skills are skipped."""
+        from praisonaiagents.skills.discovery import discover_skills
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            # Create valid skill
+            valid_dir = Path(tmpdir) / "valid-skill"
+            valid_dir.mkdir()
+            (valid_dir / "SKILL.md").write_text("""---
+name: valid-skill
+description: A valid skill
+---
+""")
+            
+            # Create invalid skill (no SKILL.md)
+            invalid_dir = Path(tmpdir) / "invalid-skill"
+            invalid_dir.mkdir()
+            
+            result = discover_skills([tmpdir], include_defaults=False)
+            
+            assert len(result) == 1
+            assert result[0].name == "valid-skill"
+    
+    def test_discover_from_multiple_directories(self):
+        """Test discovering from multiple directories."""
+        from praisonaiagents.skills.discovery import discover_skills
+        
+        with tempfile.TemporaryDirectory() as tmpdir1:
+            with tempfile.TemporaryDirectory() as tmpdir2:
+                # Create skill in first directory
+                skill1_dir = Path(tmpdir1) / "skill-a"
+                skill1_dir.mkdir()
+                (skill1_dir / "SKILL.md").write_text("""---
+name: skill-a
+description: Skill A
+---
+""")
+                
+                # Create skill in second directory
+                skill2_dir = Path(tmpdir2) / "skill-b"
+                skill2_dir.mkdir()
+                (skill2_dir / "SKILL.md").write_text("""---
+name: skill-b
+description: Skill B
+---
+""")
+                
+                result = discover_skills([tmpdir1, tmpdir2], include_defaults=False)
+                
+                assert len(result) == 2
+
+
+class TestDiscoverSkill:
+    """Tests for discover_skill function (single skill)."""
+    
+    def test_discover_valid_skill(self):
+        """Test discovering a single valid skill."""
+        from praisonaiagents.skills.discovery import discover_skill
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "my-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: my-skill
+description: My skill
+---
+""")
+            
+            result = discover_skill(str(skill_dir))
+            
+            assert result is not None
+            assert result.name == "my-skill"
+    
+    def test_discover_nonexistent_path(self):
+        """Test discovering from nonexistent path."""
+        from praisonaiagents.skills.discovery import discover_skill
+        
+        result = discover_skill("/nonexistent/path")
+        
+        assert result is None
+    
+    def test_discover_directory_without_skill_md(self):
+        """Test discovering from directory without SKILL.md."""
+        from praisonaiagents.skills.discovery import discover_skill
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            result = discover_skill(tmpdir)
+            
+            assert result is None
@@ -0,0 +1,243 @@
+"""Tests for Agent Skills loader."""
+
+from pathlib import Path
+import tempfile
+
+
+class TestSkillLoader:
+    """Tests for SkillLoader class."""
+    
+    def test_load_metadata_valid_skill(self):
+        """Test loading metadata from valid skill."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+
+# Instructions
+""")
+            
+            loader = SkillLoader()
+            skill = loader.load_metadata(str(skill_dir))
+            
+            assert skill is not None
+            assert skill.properties.name == "test-skill"
+            assert skill.instructions is None  # Not activated yet
+    
+    def test_load_metadata_nonexistent(self):
+        """Test loading from nonexistent path."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        loader = SkillLoader()
+        skill = loader.load_metadata("/nonexistent/path")
+        
+        assert skill is None
+    
+    def test_activate_skill(self):
+        """Test activating a skill loads instructions."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+
+# Instructions
+
+Follow these steps:
+1. Do this
+2. Do that
+""")
+            
+            loader = SkillLoader()
+            skill = loader.load_metadata(str(skill_dir))
+            
+            assert not skill.is_activated
+            
+            result = loader.activate(skill)
+            
+            assert result is True
+            assert skill.is_activated
+            assert "Follow these steps" in skill.instructions
+    
+    def test_activate_already_activated(self):
+        """Test activating already activated skill returns True."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+
+Instructions.
+""")
+            
+            loader = SkillLoader()
+            skill = loader.load_metadata(str(skill_dir))
+            loader.activate(skill)
+            
+            # Activate again
+            result = loader.activate(skill)
+            
+            assert result is True
+    
+    def test_load_scripts(self):
+        """Test loading scripts from skill."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            # Create scripts directory
+            scripts_dir = skill_dir / "scripts"
+            scripts_dir.mkdir()
+            (scripts_dir / "helper.py").write_text("print('hello')")
+            (scripts_dir / "utils.sh").write_text("echo 'hello'")
+            
+            loader = SkillLoader()
+            skill = loader.load_metadata(str(skill_dir))
+            scripts = loader.load_scripts(skill)
+            
+            assert "helper.py" in scripts
+            assert "utils.sh" in scripts
+            assert "print('hello')" in scripts["helper.py"]
+    
+    def test_load_references(self):
+        """Test loading references from skill."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            # Create references directory
+            refs_dir = skill_dir / "references"
+            refs_dir.mkdir()
+            (refs_dir / "api.md").write_text("# API Reference")
+            
+            loader = SkillLoader()
+            skill = loader.load_metadata(str(skill_dir))
+            refs = loader.load_references(skill)
+            
+            assert "api.md" in refs
+            assert "# API Reference" in refs["api.md"]
+    
+    def test_load_assets(self):
+        """Test loading assets from skill."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            # Create assets directory
+            assets_dir = skill_dir / "assets"
+            assets_dir.mkdir()
+            (assets_dir / "template.json").write_text("{}")
+            
+            loader = SkillLoader()
+            skill = loader.load_metadata(str(skill_dir))
+            assets = loader.load_assets(skill)
+            
+            assert "template.json" in assets
+            # Path may be resolved differently on macOS (/var vs /private/var)
+            assert assets["template.json"].endswith("test-skill/assets/template.json")
+    
+    def test_class_method_load(self):
+        """Test convenience class method load."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+
+Instructions here.
+""")
+            
+            skill = SkillLoader.load(str(skill_dir), activate=True)
+            
+            assert skill is not None
+            assert skill.is_activated
+            assert "Instructions here" in skill.instructions
+
+
+class TestLoadedSkill:
+    """Tests for LoadedSkill dataclass."""
+    
+    def test_metadata_property(self):
+        """Test metadata property returns SkillMetadata."""
+        from praisonaiagents.skills.loader import SkillLoader
+        from praisonaiagents.skills.models import SkillMetadata
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            loader = SkillLoader()
+            skill = loader.load_metadata(str(skill_dir))
+            
+            meta = skill.metadata
+            
+            assert isinstance(meta, SkillMetadata)
+            assert meta.name == "test-skill"
+            assert meta.description == "A test skill"
+    
+    def test_is_activated_property(self):
+        """Test is_activated property."""
+        from praisonaiagents.skills.loader import SkillLoader
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+
+Body.
+""")
+            
+            loader = SkillLoader()
+            skill = loader.load_metadata(str(skill_dir))
+            
+            assert skill.is_activated is False
+            
+            loader.activate(skill)
+            
+            assert skill.is_activated is True
@@ -0,0 +1,254 @@
+"""Tests for Agent Skills manager."""
+
+from pathlib import Path
+import tempfile
+
+
+class TestSkillManager:
+    """Tests for SkillManager class."""
+    
+    def test_init_empty(self):
+        """Test initializing empty manager."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        manager = SkillManager()
+        
+        assert len(manager) == 0
+        assert manager.skill_names == []
+    
+    def test_discover_skills(self):
+        """Test discovering skills from directory."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            # Create a skill
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            manager = SkillManager()
+            count = manager.discover([tmpdir], include_defaults=False)
+            
+            assert count == 1
+            assert len(manager) == 1
+            assert "test-skill" in manager
+    
+    def test_add_skill(self):
+        """Test adding a single skill."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "my-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: my-skill
+description: My skill
+---
+""")
+            
+            manager = SkillManager()
+            skill = manager.add_skill(str(skill_dir))
+            
+            assert skill is not None
+            assert skill.properties.name == "my-skill"
+            assert "my-skill" in manager
+    
+    def test_get_skill(self):
+        """Test getting a skill by name."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            manager = SkillManager()
+            manager.add_skill(str(skill_dir))
+            
+            skill = manager.get_skill("test-skill")
+            
+            assert skill is not None
+            assert skill.properties.name == "test-skill"
+    
+    def test_get_skill_not_found(self):
+        """Test getting nonexistent skill."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        manager = SkillManager()
+        skill = manager.get_skill("nonexistent")
+        
+        assert skill is None
+    
+    def test_activate_by_name(self):
+        """Test activating skill by name."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+
+# Instructions
+""")
+            
+            manager = SkillManager()
+            manager.add_skill(str(skill_dir))
+            
+            result = manager.activate_by_name("test-skill")
+            
+            assert result is True
+            skill = manager.get_skill("test-skill")
+            assert skill.is_activated
+    
+    def test_get_available_skills(self):
+        """Test getting available skills metadata."""
+        from praisonaiagents.skills.manager import SkillManager
+        from praisonaiagents.skills.models import SkillMetadata
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            manager = SkillManager()
+            manager.add_skill(str(skill_dir))
+            
+            skills = manager.get_available_skills()
+            
+            assert len(skills) == 1
+            assert isinstance(skills[0], SkillMetadata)
+            assert skills[0].name == "test-skill"
+    
+    def test_to_prompt(self):
+        """Test generating prompt XML."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            manager = SkillManager()
+            manager.add_skill(str(skill_dir))
+            
+            prompt = manager.to_prompt()
+            
+            assert "<available_skills>" in prompt
+            assert "test-skill" in prompt
+            assert "A test skill" in prompt
+    
+    def test_get_instructions(self):
+        """Test getting instructions for a skill."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+
+# Instructions
+
+Do this and that.
+""")
+            
+            manager = SkillManager()
+            manager.add_skill(str(skill_dir))
+            
+            instructions = manager.get_instructions("test-skill")
+            
+            assert instructions is not None
+            assert "Do this and that" in instructions
+    
+    def test_load_resources(self):
+        """Test loading resources for a skill."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            # Create scripts directory
+            scripts_dir = skill_dir / "scripts"
+            scripts_dir.mkdir()
+            (scripts_dir / "helper.py").write_text("print('hello')")
+            
+            manager = SkillManager()
+            manager.add_skill(str(skill_dir))
+            
+            result = manager.load_resources("test-skill")
+            
+            assert result is True
+            skill = manager.get_skill("test-skill")
+            assert skill.resources_loaded
+    
+    def test_clear(self):
+        """Test clearing all skills."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill
+---
+""")
+            
+            manager = SkillManager()
+            manager.add_skill(str(skill_dir))
+            assert len(manager) == 1
+            
+            manager.clear()
+            
+            assert len(manager) == 0
+    
+    def test_iteration(self):
+        """Test iterating over skills."""
+        from praisonaiagents.skills.manager import SkillManager
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            # Create two skills
+            for name in ["skill-a", "skill-b"]:
+                skill_dir = Path(tmpdir) / name
+                skill_dir.mkdir()
+                (skill_dir / "SKILL.md").write_text(f"""---
+name: {name}
+description: Skill {name}
+---
+""")
+            
+            manager = SkillManager()
+            manager.discover([tmpdir], include_defaults=False)
+            
+            names = [s.properties.name for s in manager]
+            
+            assert len(names) == 2
+            assert "skill-a" in names
+            assert "skill-b" in names
@@ -0,0 +1,184 @@
+"""Tests for Agent Skills models."""
+
+from pathlib import Path
+
+
+class TestSkillProperties:
+    """Tests for SkillProperties dataclass."""
+    
+    def test_skill_properties_required_fields(self):
+        """Test that SkillProperties requires name and description."""
+        from praisonaiagents.skills.models import SkillProperties
+        
+        skill = SkillProperties(
+            name="test-skill",
+            description="A test skill for unit testing"
+        )
+        
+        assert skill.name == "test-skill"
+        assert skill.description == "A test skill for unit testing"
+    
+    def test_skill_properties_optional_fields_default_none(self):
+        """Test that optional fields default to None."""
+        from praisonaiagents.skills.models import SkillProperties
+        
+        skill = SkillProperties(
+            name="test-skill",
+            description="A test skill"
+        )
+        
+        assert skill.license is None
+        assert skill.compatibility is None
+        assert skill.allowed_tools is None
+        assert skill.path is None
+    
+    def test_skill_properties_metadata_defaults_empty_dict(self):
+        """Test that metadata defaults to empty dict."""
+        from praisonaiagents.skills.models import SkillProperties
+        
+        skill = SkillProperties(
+            name="test-skill",
+            description="A test skill"
+        )
+        
+        assert skill.metadata == {}
+    
+    def test_skill_properties_all_fields(self):
+        """Test SkillProperties with all fields populated."""
+        from praisonaiagents.skills.models import SkillProperties
+        
+        skill = SkillProperties(
+            name="pdf-processing",
+            description="Extract text and tables from PDF files",
+            license="Apache-2.0",
+            compatibility="Requires pypdf package",
+            allowed_tools="Read Grep Bash",
+            metadata={"author": "test-org", "version": "1.0"},
+            path=Path("/path/to/skill")
+        )
+        
+        assert skill.name == "pdf-processing"
+        assert skill.description == "Extract text and tables from PDF files"
+        assert skill.license == "Apache-2.0"
+        assert skill.compatibility == "Requires pypdf package"
+        assert skill.allowed_tools == "Read Grep Bash"
+        assert skill.metadata == {"author": "test-org", "version": "1.0"}
+        assert skill.path == Path("/path/to/skill")
+    
+    def test_skill_properties_to_dict(self):
+        """Test to_dict method excludes None values."""
+        from praisonaiagents.skills.models import SkillProperties
+        
+        skill = SkillProperties(
+            name="test-skill",
+            description="A test skill",
+            license="MIT"
+        )
+        
+        result = skill.to_dict()
+        
+        assert result["name"] == "test-skill"
+        assert result["description"] == "A test skill"
+        assert result["license"] == "MIT"
+        assert "compatibility" not in result
+        assert "allowed-tools" not in result
+        assert "metadata" not in result  # Empty metadata excluded
+    
+    def test_skill_properties_to_dict_with_metadata(self):
+        """Test to_dict includes non-empty metadata."""
+        from praisonaiagents.skills.models import SkillProperties
+        
+        skill = SkillProperties(
+            name="test-skill",
+            description="A test skill",
+            metadata={"version": "1.0"}
+        )
+        
+        result = skill.to_dict()
+        
+        assert result["metadata"] == {"version": "1.0"}
+    
+    def test_skill_properties_to_dict_allowed_tools_key(self):
+        """Test to_dict uses 'allowed-tools' key (hyphenated)."""
+        from praisonaiagents.skills.models import SkillProperties
+        
+        skill = SkillProperties(
+            name="test-skill",
+            description="A test skill",
+            allowed_tools="Read Write"
+        )
+        
+        result = skill.to_dict()
+        
+        assert "allowed-tools" in result
+        assert result["allowed-tools"] == "Read Write"
+
+
+class TestSkillMetadata:
+    """Tests for SkillMetadata dataclass (lightweight for system prompt)."""
+    
+    def test_skill_metadata_required_fields(self):
+        """Test SkillMetadata with required fields."""
+        from praisonaiagents.skills.models import SkillMetadata
+        
+        meta = SkillMetadata(
+            name="test-skill",
+            description="A test skill",
+            location="/path/to/skill/SKILL.md"
+        )
+        
+        assert meta.name == "test-skill"
+        assert meta.description == "A test skill"
+        assert meta.location == "/path/to/skill/SKILL.md"
+    
+    def test_skill_metadata_from_properties(self):
+        """Test creating SkillMetadata from SkillProperties."""
+        from praisonaiagents.skills.models import SkillProperties, SkillMetadata
+        
+        props = SkillProperties(
+            name="pdf-skill",
+            description="Process PDF files",
+            path=Path("/skills/pdf-skill")
+        )
+        
+        meta = SkillMetadata.from_properties(props)
+        
+        assert meta.name == "pdf-skill"
+        assert meta.description == "Process PDF files"
+        assert "/skills/pdf-skill" in meta.location
+
+
+class TestParseError:
+    """Tests for ParseError exception."""
+    
+    def test_parse_error_message(self):
+        """Test ParseError stores message correctly."""
+        from praisonaiagents.skills.models import ParseError
+        
+        error = ParseError("Invalid YAML frontmatter")
+        
+        assert str(error) == "Invalid YAML frontmatter"
+    
+    def test_parse_error_is_exception(self):
+        """Test ParseError is an Exception subclass."""
+        from praisonaiagents.skills.models import ParseError
+        
+        assert issubclass(ParseError, Exception)
+
+
+class TestValidationError:
+    """Tests for ValidationError exception."""
+    
+    def test_validation_error_message(self):
+        """Test ValidationError stores message correctly."""
+        from praisonaiagents.skills.models import ValidationError
+        
+        error = ValidationError("Missing required field: name")
+        
+        assert str(error) == "Missing required field: name"
+    
+    def test_validation_error_is_exception(self):
+        """Test ValidationError is an Exception subclass."""
+        from praisonaiagents.skills.models import ValidationError
+        
+        assert issubclass(ValidationError, Exception)
@@ -0,0 +1,288 @@
+"""Tests for Agent Skills parser."""
+
+from pathlib import Path
+import tempfile
+
+
+class TestParseFrontmatter:
+    """Tests for parse_frontmatter function."""
+    
+    def test_parse_valid_frontmatter(self):
+        """Test parsing valid YAML frontmatter."""
+        from praisonaiagents.skills.parser import parse_frontmatter
+        
+        content = """---
+name: test-skill
+description: A test skill for unit testing
+---
+
+# Test Skill
+
+Instructions go here.
+"""
+        metadata, body = parse_frontmatter(content)
+        
+        assert metadata["name"] == "test-skill"
+        assert metadata["description"] == "A test skill for unit testing"
+        assert "# Test Skill" in body
+    
+    def test_parse_frontmatter_with_optional_fields(self):
+        """Test parsing frontmatter with all optional fields."""
+        from praisonaiagents.skills.parser import parse_frontmatter
+        
+        content = """---
+name: pdf-processing
+description: Extract text from PDFs
+license: Apache-2.0
+compatibility: Requires pypdf
+allowed-tools: Read Grep
+metadata:
+  author: test-org
+  version: "1.0"
+---
+
+Body content.
+"""
+        metadata, body = parse_frontmatter(content)
+        
+        assert metadata["name"] == "pdf-processing"
+        assert metadata["license"] == "Apache-2.0"
+        assert metadata["compatibility"] == "Requires pypdf"
+        assert metadata["allowed-tools"] == "Read Grep"
+        assert metadata["metadata"]["author"] == "test-org"
+        assert metadata["metadata"]["version"] == "1.0"
+    
+    def test_parse_frontmatter_missing_opening_delimiter(self):
+        """Test error when frontmatter doesn't start with ---."""
+        from praisonaiagents.skills.parser import parse_frontmatter
+        from praisonaiagents.skills.models import ParseError
+        
+        content = """name: test-skill
+description: Missing opening delimiter
+---
+
+Body content.
+"""
+        try:
+            parse_frontmatter(content)
+            assert False, "Should have raised ParseError"
+        except ParseError as e:
+            assert "must start with" in str(e).lower() or "---" in str(e)
+    
+    def test_parse_frontmatter_missing_closing_delimiter(self):
+        """Test error when frontmatter is not closed."""
+        from praisonaiagents.skills.parser import parse_frontmatter
+        from praisonaiagents.skills.models import ParseError
+        
+        content = """---
+name: test-skill
+description: Missing closing delimiter
+
+Body content without closing delimiter.
+"""
+        try:
+            parse_frontmatter(content)
+            assert False, "Should have raised ParseError"
+        except ParseError as e:
+            assert "closed" in str(e).lower() or "---" in str(e)
+    
+    def test_parse_frontmatter_invalid_yaml(self):
+        """Test error when YAML is invalid."""
+        from praisonaiagents.skills.parser import parse_frontmatter
+        from praisonaiagents.skills.models import ParseError
+        
+        content = """---
+name: test-skill
+description: [invalid yaml
+  - missing bracket
+---
+
+Body content.
+"""
+        try:
+            parse_frontmatter(content)
+            assert False, "Should have raised ParseError"
+        except ParseError as e:
+            assert "yaml" in str(e).lower() or "invalid" in str(e).lower()
+    
+    def test_parse_frontmatter_empty_body(self):
+        """Test parsing when body is empty."""
+        from praisonaiagents.skills.parser import parse_frontmatter
+        
+        content = """---
+name: test-skill
+description: A skill with no body
+---
+"""
+        metadata, body = parse_frontmatter(content)
+        
+        assert metadata["name"] == "test-skill"
+        assert body.strip() == ""
+
+
+class TestFindSkillMd:
+    """Tests for find_skill_md function."""
+    
+    def test_find_skill_md_uppercase(self):
+        """Test finding SKILL.md (uppercase)."""
+        from praisonaiagents.skills.parser import find_skill_md
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir)
+            skill_md = skill_dir / "SKILL.md"
+            skill_md.write_text("---\nname: test\ndescription: test\n---\n")
+            
+            result = find_skill_md(skill_dir)
+            
+            assert result is not None
+            assert result.name == "SKILL.md"
+    
+    def test_find_skill_md_lowercase(self):
+        """Test finding skill.md (lowercase) as fallback."""
+        from praisonaiagents.skills.parser import find_skill_md
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir)
+            skill_md = skill_dir / "skill.md"
+            skill_md.write_text("---\nname: test\ndescription: test\n---\n")
+            
+            result = find_skill_md(skill_dir)
+            
+            assert result is not None
+            # On case-insensitive filesystems (macOS), SKILL.md may match skill.md
+            assert result.name.lower() == "skill.md"
+    
+    def test_find_skill_md_prefers_uppercase(self):
+        """Test that SKILL.md is preferred over skill.md."""
+        from praisonaiagents.skills.parser import find_skill_md
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir)
+            (skill_dir / "SKILL.md").write_text("uppercase")
+            (skill_dir / "skill.md").write_text("lowercase")
+            
+            result = find_skill_md(skill_dir)
+            
+            assert result is not None
+            assert result.name == "SKILL.md"
+    
+    def test_find_skill_md_not_found(self):
+        """Test returning None when no SKILL.md exists."""
+        from praisonaiagents.skills.parser import find_skill_md
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir)
+            
+            result = find_skill_md(skill_dir)
+            
+            assert result is None
+
+
+class TestReadProperties:
+    """Tests for read_properties function."""
+    
+    def test_read_properties_valid_skill(self):
+        """Test reading properties from a valid skill directory."""
+        from praisonaiagents.skills.parser import read_properties
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill for unit testing
+---
+
+# Test Skill
+""")
+            
+            props = read_properties(skill_dir)
+            
+            assert props.name == "test-skill"
+            assert props.description == "A test skill for unit testing"
+    
+    def test_read_properties_with_all_fields(self):
+        """Test reading properties with all optional fields."""
+        from praisonaiagents.skills.parser import read_properties
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "pdf-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: pdf-skill
+description: Process PDF files
+license: MIT
+compatibility: Requires pypdf
+allowed-tools: Read Write
+metadata:
+  author: test
+  version: "2.0"
+---
+
+Instructions.
+""")
+            
+            props = read_properties(skill_dir)
+            
+            assert props.name == "pdf-skill"
+            assert props.license == "MIT"
+            assert props.compatibility == "Requires pypdf"
+            assert props.allowed_tools == "Read Write"
+            assert props.metadata["author"] == "test"
+    
+    def test_read_properties_missing_skill_md(self):
+        """Test error when SKILL.md is missing."""
+        from praisonaiagents.skills.parser import read_properties
+        from praisonaiagents.skills.models import ParseError
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "empty-skill"
+            skill_dir.mkdir()
+            
+            try:
+                read_properties(skill_dir)
+                assert False, "Should have raised ParseError"
+            except ParseError as e:
+                assert "not found" in str(e).lower() or "skill.md" in str(e).lower()
+    
+    def test_read_properties_missing_name(self):
+        """Test error when name field is missing."""
+        from praisonaiagents.skills.parser import read_properties
+        from praisonaiagents.skills.models import ValidationError
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "no-name"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+description: Missing name field
+---
+
+Body.
+""")
+            
+            try:
+                read_properties(skill_dir)
+                assert False, "Should have raised ValidationError"
+            except ValidationError as e:
+                assert "name" in str(e).lower()
+    
+    def test_read_properties_missing_description(self):
+        """Test error when description field is missing."""
+        from praisonaiagents.skills.parser import read_properties
+        from praisonaiagents.skills.models import ValidationError
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "no-desc"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: no-desc
+---
+
+Body.
+""")
+            
+            try:
+                read_properties(skill_dir)
+                assert False, "Should have raised ValidationError"
+            except ValidationError as e:
+                assert "description" in str(e).lower()
@@ -0,0 +1,180 @@
+"""Tests for Agent Skills prompt generation."""
+
+from pathlib import Path
+import tempfile
+
+
+class TestToPrompt:
+    """Tests for to_prompt function (XML generation)."""
+    
+    def test_to_prompt_empty_list(self):
+        """Test generating prompt with no skills."""
+        from praisonaiagents.skills.prompt import to_prompt
+        
+        result = to_prompt([])
+        
+        assert "<available_skills>" in result
+        assert "</available_skills>" in result
+    
+    def test_to_prompt_single_skill(self):
+        """Test generating prompt with one skill."""
+        from praisonaiagents.skills.prompt import to_prompt
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill for prompt generation
+---
+
+# Test Skill
+""")
+            
+            result = to_prompt([skill_dir])
+            
+            assert "<available_skills>" in result
+            assert "</available_skills>" in result
+            assert "<skill>" in result
+            assert "</skill>" in result
+            assert "<name>" in result
+            assert "test-skill" in result
+            assert "<description>" in result
+            assert "A test skill for prompt generation" in result
+            assert "<location>" in result
+    
+    def test_to_prompt_multiple_skills(self):
+        """Test generating prompt with multiple skills."""
+        from praisonaiagents.skills.prompt import to_prompt
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            # Create first skill
+            skill1_dir = Path(tmpdir) / "skill-one"
+            skill1_dir.mkdir()
+            (skill1_dir / "SKILL.md").write_text("""---
+name: skill-one
+description: First skill
+---
+""")
+            
+            # Create second skill
+            skill2_dir = Path(tmpdir) / "skill-two"
+            skill2_dir.mkdir()
+            (skill2_dir / "SKILL.md").write_text("""---
+name: skill-two
+description: Second skill
+---
+""")
+            
+            result = to_prompt([skill1_dir, skill2_dir])
+            
+            assert result.count("<skill>") == 2
+            assert result.count("</skill>") == 2
+            assert "skill-one" in result
+            assert "skill-two" in result
+            assert "First skill" in result
+            assert "Second skill" in result
+    
+    def test_to_prompt_escapes_html(self):
+        """Test that special characters are HTML escaped."""
+        from praisonaiagents.skills.prompt import to_prompt
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "escape-test"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: escape-test
+description: Test <script> & "quotes" handling
+---
+""")
+            
+            result = to_prompt([skill_dir])
+            
+            # Should escape < > & "
+            assert "<script>" not in result or "&lt;script&gt;" in result
+            assert "&amp;" in result or "& " not in result
+    
+    def test_to_prompt_includes_location(self):
+        """Test that location points to SKILL.md file."""
+        from praisonaiagents.skills.prompt import to_prompt
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "location-test"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: location-test
+description: Test location field
+---
+""")
+            
+            result = to_prompt([skill_dir])
+            
+            assert "<location>" in result
+            assert "SKILL.md" in result
+            assert "</location>" in result
+
+
+class TestGenerateSkillsXml:
+    """Tests for generate_skills_xml helper function."""
+    
+    def test_generate_skills_xml_from_metadata_list(self):
+        """Test generating XML from SkillMetadata list."""
+        from praisonaiagents.skills.prompt import generate_skills_xml
+        from praisonaiagents.skills.models import SkillMetadata
+        
+        skills = [
+            SkillMetadata(
+                name="pdf-skill",
+                description="Process PDF files",
+                location="/path/to/pdf-skill/SKILL.md"
+            ),
+            SkillMetadata(
+                name="excel-skill",
+                description="Process Excel files",
+                location="/path/to/excel-skill/SKILL.md"
+            )
+        ]
+        
+        result = generate_skills_xml(skills)
+        
+        assert "<available_skills>" in result
+        assert "pdf-skill" in result
+        assert "excel-skill" in result
+        assert "Process PDF files" in result
+        assert "Process Excel files" in result
+    
+    def test_generate_skills_xml_empty_list(self):
+        """Test generating XML with empty list."""
+        from praisonaiagents.skills.prompt import generate_skills_xml
+        
+        result = generate_skills_xml([])
+        
+        assert "<available_skills>" in result
+        assert "</available_skills>" in result
+        assert "<skill>" not in result
+
+
+class TestFormatSkillForPrompt:
+    """Tests for format_skill_for_prompt helper."""
+    
+    def test_format_single_skill(self):
+        """Test formatting a single skill for prompt."""
+        from praisonaiagents.skills.prompt import format_skill_for_prompt
+        from praisonaiagents.skills.models import SkillMetadata
+        
+        skill = SkillMetadata(
+            name="test-skill",
+            description="A test skill",
+            location="/path/to/SKILL.md"
+        )
+        
+        result = format_skill_for_prompt(skill)
+        
+        assert "<skill>" in result
+        assert "</skill>" in result
+        assert "<name>" in result
+        assert "test-skill" in result
+        assert "<description>" in result
+        assert "A test skill" in result
+        assert "<location>" in result
+        assert "/path/to/SKILL.md" in result
@@ -0,0 +1,258 @@
+"""Tests for Agent Skills validator."""
+
+from pathlib import Path
+import tempfile
+
+
+class TestValidateName:
+    """Tests for skill name validation."""
+    
+    def test_valid_name_simple(self):
+        """Test valid simple skill name."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        errors = _validate_name("test-skill", None)
+        
+        assert errors == []
+    
+    def test_valid_name_with_numbers(self):
+        """Test valid name with numbers."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        errors = _validate_name("skill-v2", None)
+        
+        assert errors == []
+    
+    def test_invalid_name_uppercase(self):
+        """Test that uppercase letters are rejected."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        errors = _validate_name("Test-Skill", None)
+        
+        assert len(errors) > 0
+        assert any("lowercase" in e.lower() for e in errors)
+    
+    def test_invalid_name_starts_with_hyphen(self):
+        """Test that names starting with hyphen are rejected."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        errors = _validate_name("-skill", None)
+        
+        assert len(errors) > 0
+        assert any("hyphen" in e.lower() for e in errors)
+    
+    def test_invalid_name_ends_with_hyphen(self):
+        """Test that names ending with hyphen are rejected."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        errors = _validate_name("skill-", None)
+        
+        assert len(errors) > 0
+        assert any("hyphen" in e.lower() for e in errors)
+    
+    def test_invalid_name_consecutive_hyphens(self):
+        """Test that consecutive hyphens are rejected."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        errors = _validate_name("test--skill", None)
+        
+        assert len(errors) > 0
+        assert any("consecutive" in e.lower() for e in errors)
+    
+    def test_invalid_name_too_long(self):
+        """Test that names over 64 characters are rejected."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        long_name = "a" * 65
+        errors = _validate_name(long_name, None)
+        
+        assert len(errors) > 0
+        assert any("64" in e or "limit" in e.lower() for e in errors)
+    
+    def test_invalid_name_special_characters(self):
+        """Test that special characters are rejected."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        errors = _validate_name("test_skill", None)
+        
+        assert len(errors) > 0
+        assert any("invalid" in e.lower() or "character" in e.lower() for e in errors)
+    
+    def test_name_must_match_directory(self):
+        """Test that name must match parent directory name."""
+        from praisonaiagents.skills.validator import _validate_name
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "actual-name"
+            skill_dir.mkdir()
+            
+            errors = _validate_name("different-name", skill_dir)
+            
+            assert len(errors) > 0
+            assert any("match" in e.lower() for e in errors)
+
+
+class TestValidateDescription:
+    """Tests for skill description validation."""
+    
+    def test_valid_description(self):
+        """Test valid description."""
+        from praisonaiagents.skills.validator import _validate_description
+        
+        errors = _validate_description("A helpful skill that does things")
+        
+        assert errors == []
+    
+    def test_invalid_description_empty(self):
+        """Test that empty description is rejected."""
+        from praisonaiagents.skills.validator import _validate_description
+        
+        errors = _validate_description("")
+        
+        assert len(errors) > 0
+        assert any("empty" in e.lower() or "non-empty" in e.lower() for e in errors)
+    
+    def test_invalid_description_too_long(self):
+        """Test that descriptions over 1024 characters are rejected."""
+        from praisonaiagents.skills.validator import _validate_description
+        
+        long_desc = "a" * 1025
+        errors = _validate_description(long_desc)
+        
+        assert len(errors) > 0
+        assert any("1024" in e or "limit" in e.lower() for e in errors)
+
+
+class TestValidateCompatibility:
+    """Tests for skill compatibility validation."""
+    
+    def test_valid_compatibility(self):
+        """Test valid compatibility string."""
+        from praisonaiagents.skills.validator import _validate_compatibility
+        
+        errors = _validate_compatibility("Requires Python 3.11+")
+        
+        assert errors == []
+    
+    def test_invalid_compatibility_too_long(self):
+        """Test that compatibility over 500 characters is rejected."""
+        from praisonaiagents.skills.validator import _validate_compatibility
+        
+        long_compat = "a" * 501
+        errors = _validate_compatibility(long_compat)
+        
+        assert len(errors) > 0
+        assert any("500" in e or "limit" in e.lower() for e in errors)
+
+
+class TestValidateMetadata:
+    """Tests for full metadata validation."""
+    
+    def test_valid_metadata(self):
+        """Test valid complete metadata."""
+        from praisonaiagents.skills.validator import validate_metadata
+        
+        metadata = {
+            "name": "test-skill",
+            "description": "A test skill"
+        }
+        
+        errors = validate_metadata(metadata)
+        
+        assert errors == []
+    
+    def test_missing_name(self):
+        """Test error when name is missing."""
+        from praisonaiagents.skills.validator import validate_metadata
+        
+        metadata = {
+            "description": "A test skill"
+        }
+        
+        errors = validate_metadata(metadata)
+        
+        assert len(errors) > 0
+        assert any("name" in e.lower() for e in errors)
+    
+    def test_missing_description(self):
+        """Test error when description is missing."""
+        from praisonaiagents.skills.validator import validate_metadata
+        
+        metadata = {
+            "name": "test-skill"
+        }
+        
+        errors = validate_metadata(metadata)
+        
+        assert len(errors) > 0
+        assert any("description" in e.lower() for e in errors)
+    
+    def test_unexpected_fields_rejected(self):
+        """Test that unexpected fields are rejected."""
+        from praisonaiagents.skills.validator import validate_metadata
+        
+        metadata = {
+            "name": "test-skill",
+            "description": "A test skill",
+            "unknown-field": "value"
+        }
+        
+        errors = validate_metadata(metadata)
+        
+        assert len(errors) > 0
+        assert any("unexpected" in e.lower() or "unknown-field" in e for e in errors)
+
+
+class TestValidate:
+    """Tests for full skill directory validation."""
+    
+    def test_validate_valid_skill(self):
+        """Test validating a valid skill directory."""
+        from praisonaiagents.skills.validator import validate
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "test-skill"
+            skill_dir.mkdir()
+            (skill_dir / "SKILL.md").write_text("""---
+name: test-skill
+description: A test skill for validation
+---
+
+# Test Skill
+""")
+            
+            errors = validate(skill_dir)
+            
+            assert errors == []
+    
+    def test_validate_missing_skill_md(self):
+        """Test error when SKILL.md is missing."""
+        from praisonaiagents.skills.validator import validate
+        
+        with tempfile.TemporaryDirectory() as tmpdir:
+            skill_dir = Path(tmpdir) / "empty-skill"
+            skill_dir.mkdir()
+            
+            errors = validate(skill_dir)
+            
+            assert len(errors) > 0
+            assert any("skill.md" in e.lower() for e in errors)
+    
+    def test_validate_nonexistent_directory(self):
+        """Test error when directory doesn't exist."""
+        from praisonaiagents.skills.validator import validate
+        
+        errors = validate(Path("/nonexistent/path"))
+        
+        assert len(errors) > 0
+        assert any("exist" in e.lower() for e in errors)
+    
+    def test_validate_not_a_directory(self):
+        """Test error when path is not a directory."""
+        from praisonaiagents.skills.validator import validate
+        
+        with tempfile.NamedTemporaryFile() as tmpfile:
+            errors = validate(Path(tmpfile.name))
+            
+            assert len(errors) > 0
+            assert any("directory" in e.lower() for e in errors)
@@ -2422,7 +2422,7 @@ wheels = [

 [[package]]
 name = "praisonaiagents"
-version = "0.1.17"
+version = "0.1.23"
 source = { editable = "." }
 dependencies = [
    { name = "aiohttp" },
@@ -6,10 +6,13 @@
  </picture>
 </p>

+<!-- mcp-name: io.github.MervinPraison/praisonai -->
+
 <p align="center">
 <a href="https://github.com/MervinPraison/PraisonAI"><img src="https://static.pepy.tech/badge/PraisonAI" alt="Total Downloads" /></a>
 <a href="https://github.com/MervinPraison/PraisonAI"><img src="https://img.shields.io/github/v/release/MervinPraison/PraisonAI" alt="Latest Stable Version" /></a>
 <a href="https://github.com/MervinPraison/PraisonAI"><img src="https://img.shields.io/badge/License-MIT-yellow.svg" alt="License" /></a>
+<a href="https://registry.modelcontextprotocol.io/servers/io.github.MervinPraison/praisonai"><img src="https://img.shields.io/badge/MCP-Registry-blue" alt="MCP Registry" /></a>
 </p>

 <div align="center">
@@ -168,12 +171,13 @@ npm install praisonai

 | Feature | Code | Docs |
 |---------|:----:|:----:|
-| 🚀 Single Agent | [Example](examples/python/agents/single-agent.py) | [📖](https://docs.praison.ai/agents/single) |
-| 🤝 Multi Agents | [Example](examples/python/general/mini_agents_example.py) | [📖](https://docs.praison.ai/concepts/agents) |
-| 🤖 Auto Agents | [Example](examples/python/general/auto_agents_example.py) | [📖](https://docs.praison.ai/features/autoagents) |
-| 🔄 Self Reflection AI Agents | [Example](examples/python/concepts/self-reflection-details.py) | [📖](https://docs.praison.ai/features/selfreflection) |
-| 🧠 Reasoning AI Agents | [Example](examples/python/concepts/reasoning-extraction.py) | [📖](https://docs.praison.ai/features/reasoning) |
-| 👁️ Multi Modal AI Agents | [Example](examples/python/general/multimodal.py) | [📖](https://docs.praison.ai/features/multimodal) |
+| **🤖 Core Agents** | | |
+| ↳ Single Agent | [Example](examples/python/agents/single-agent.py) | [📖](https://docs.praison.ai/agents/single) |
+| ↳ Multi Agents | [Example](examples/python/general/mini_agents_example.py) | [📖](https://docs.praison.ai/concepts/agents) |
+| ↳ Auto Agents | [Example](examples/python/general/auto_agents_example.py) | [📖](https://docs.praison.ai/features/autoagents) |
+| ↳ Self Reflection AI Agents | [Example](examples/python/concepts/self-reflection-details.py) | [📖](https://docs.praison.ai/features/selfreflection) |
+| ↳ Reasoning AI Agents | [Example](examples/python/concepts/reasoning-extraction.py) | [📖](https://docs.praison.ai/features/reasoning) |
+| ↳ Multi Modal AI Agents | [Example](examples/python/general/multimodal.py) | [📖](https://docs.praison.ai/features/multimodal) |
 | **🔄 Workflows** | | |
 | ↳ Simple Workflow | [Example](examples/python/workflows/simple_workflow.py) | [📖](https://docs.praison.ai/features/workflows) |
 | ↳ Workflow with Agents | [Example](examples/python/workflows/workflow_with_agents.py) | [📖](https://docs.praison.ai/features/workflows) |
@@ -185,69 +189,105 @@ npm install praisonai
 | ↳ Workflow Branching | [Example](examples/python/workflows/workflow_branching.py) | [📖](https://docs.praison.ai/features/workflows) |
 | ↳ Workflow Early Stop | [Example](examples/python/workflows/workflow_early_stop.py) | [📖](https://docs.praison.ai/features/workflows) |
 | ↳ Workflow Checkpoints | [Example](examples/python/workflows/workflow_checkpoints.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 📚 Add Custom Knowledge | [Example](examples/python/concepts/knowledge-agents.py) | [📖](https://docs.praison.ai/features/knowledge) |
-| 🧠 Memory (Short & Long Term) | [Example](examples/python/general/memory_example.py) | [📖](https://docs.praison.ai/concepts/memory) |
-| 📄 Chat with PDF Agents | [Example](examples/python/concepts/chat-with-pdf.py) | [📖](https://docs.praison.ai/features/chat-with-pdf) |
-| 💻 Code Interpreter Agents | [Example](examples/python/agents/code-agent.py) | [📖](https://docs.praison.ai/features/codeagent) |
-| ✏️ AI Code Editing Tools | [Example](#ai-code-editing) | [📖](https://docs.praison.ai/features/code) |
-| 📚 RAG Agents | [Example](examples/python/concepts/rag-agents.py) | [📖](https://docs.praison.ai/features/rag) |
-| 🤔 Async & Parallel Processing | [Example](examples/python/general/async_example.py) | [📖](https://docs.praison.ai/features/async) |
-| 🔢 Math Agents | [Example](examples/python/agents/math-agent.py) | [📖](https://docs.praison.ai/features/mathagent) |
-| 🎯 Structured Output Agents | [Example](examples/python/general/structured_agents_example.py) | [📖](https://docs.praison.ai/features/structured) |
-| 🔗 LangChain Integrated Agents | [Example](examples/python/general/langchain_example.py) | [📖](https://docs.praison.ai/features/langchain) |
-| 📞 Callback Agents | [Example](examples/python/general/advanced-callback-systems.py) | [📖](https://docs.praison.ai/features/callbacks) |
-| 🛠️ 100+ Custom Tools | [Example](examples/python/general/tools_example.py) | [📖](https://docs.praison.ai/tools/tools) |
-| 📄 YAML Configuration | [Example](examples/cookbooks/yaml/secondary_market_research_agents.yaml) | [📖](https://docs.praison.ai/developers/agents-playbook) |
-| 💯 100+ LLM Support | [Example](examples/python/providers/openai/openai_gpt4_example.py) | [📖](https://docs.praison.ai/models) |
-| 🔬 Deep Research Agents | [Example](examples/python/agents/research-agent.py) | [📖](https://docs.praison.ai/agents/deep-research) |
-| 🔄 Query Rewriter Agent | [Example](#5-query-rewriter-agent) | [📖](https://docs.praison.ai/agents/query-rewriter) |
-| 🌐 Native Web Search | [Example](examples/python/agents/websearch-agent.py) | [📖](https://docs.praison.ai/agents/websearch) |
-| 📥 Web Fetch (Anthropic) | [Example](#web-search-web-fetch--prompt-caching) | [📖](https://docs.praison.ai/features/model-capabilities) |
-| 💾 Prompt Caching | [Example](#web-search-web-fetch--prompt-caching) | [📖](https://docs.praison.ai/features/model-capabilities) |
-| 🧠 Claude Memory Tool | [Example](#claude-memory-tool-cli) | [📖](https://docs.praison.ai/features/claude-memory-tool) |
-| 💾 File-Based Memory | [Example](examples/python/general/memory_example.py) | [📖](https://docs.praison.ai/concepts/memory) |
-| 🔍 Built-in Search Tools | [Example](examples/python/agents/websearch-agent.py) | [📖](https://docs.praison.ai/tools/tavily) |
-| 📋 Planning Mode | [Example](examples/python/agents/planning-agent.py) | [📖](https://docs.praison.ai/features/planning-mode) |
-| 🔧 Planning Tools | [Example](#3-agent-with-planning-mode) | [📖](https://docs.praison.ai/features/planning-mode) |
-| 🧠 Planning Reasoning | [Example](#3-agent-with-planning-mode) | [📖](https://docs.praison.ai/features/planning-mode) |
-| 🔌 MCP Transports | [Example](examples/python/mcp/mcp-transports-overview.py) | [📖](https://docs.praison.ai/mcp/transports) |
-| 🌐 WebSocket MCP | [Example](examples/python/mcp/websocket-mcp.py) | [📖](https://docs.praison.ai/mcp/sse-transport) |
-| 🔐 MCP Security | [Example](examples/python/mcp/mcp-security.py) | [📖](https://docs.praison.ai/mcp/transports) |
-| 🔄 MCP Resumability | [Example](examples/python/mcp/mcp-resumability.py) | [📖](https://docs.praison.ai/mcp/sse-transport) |
-| ⚡ Fast Context | [Example](examples/context/00_agent_fast_context_basic.py) | [📖](https://docs.praison.ai/features/fast-context) |
-| 🖼️ Image Generation Agent | [Example](examples/python/image/image-agent.py) | [📖](https://docs.praison.ai/features/image-generation) |
-| 📷 Image to Text Agent | [Example](examples/python/agents/image-to-text-agent.py) | [📖](https://docs.praison.ai/agents/image-to-text) |
-| 🎬 Video Agent | [Example](examples/python/agents/video-agent.py) | [📖](https://docs.praison.ai/agents/video) |
-| 📊 Data Analyst Agent | [Example](examples/python/agents/data-analyst-agent.py) | [📖](https://docs.praison.ai/agents/data-analyst) |
-| 💰 Finance Agent | [Example](examples/python/agents/finance-agent.py) | [📖](https://docs.praison.ai/agents/finance) |
-| 🛒 Shopping Agent | [Example](examples/python/agents/shopping-agent.py) | [📖](https://docs.praison.ai/agents/shopping) |
-| ⭐ Recommendation Agent | [Example](examples/python/agents/recommendation-agent.py) | [📖](https://docs.praison.ai/agents/recommendation) |
-| 📖 Wikipedia Agent | [Example](examples/python/agents/wikipedia-agent.py) | [📖](https://docs.praison.ai/agents/wikipedia) |
-| 💻 Programming Agent | [Example](examples/python/agents/programming-agent.py) | [📖](https://docs.praison.ai/agents/programming) |
-| 📝 Markdown Agent | [Example](examples/python/agents/markdown-agent.py) | [📖](https://docs.praison.ai/agents/markdown) |
-| 📝 Prompt Expander Agent | [Example](#prompt-expansion) | [📖](https://docs.praison.ai/agents/prompt-expander) |
-| 🔀 Router Agent | [Example](examples/python/agents/router-agent-cost-optimization.py) | [📖](https://docs.praison.ai/features/routing) |
-| ⛓️ Prompt Chaining | [Example](examples/python/general/prompt_chaining.py) | [📖](https://docs.praison.ai/features/promptchaining) |
-| 🔍 Evaluator Optimiser | [Example](examples/python/general/evaluator-optimiser.py) | [📖](https://docs.praison.ai/features/evaluator-optimiser) |
-| 👷 Orchestrator Workers | [Example](examples/python/general/orchestrator-workers.py) | [📖](https://docs.praison.ai/features/orchestrator-worker) |
-| ⚡ Parallelisation | [Example](examples/python/general/parallelisation.py) | [📖](https://docs.praison.ai/features/parallelisation) |
-| 🔁 Repetitive Agents | [Example](examples/python/concepts/repetitive-agents.py) | [📖](https://docs.praison.ai/features/repetitive) |
-| 🤝 Agent Handoffs | [Example](examples/python/handoff/handoff_basic.py) | [📖](https://docs.praison.ai/features/handoffs) |
-| 🛡️ Guardrails | [Example](examples/python/guardrails/comprehensive-guardrails-example.py) | [📖](https://docs.praison.ai/features/guardrails) |
-| 💬 Sessions Management | [Example](examples/python/sessions/comprehensive-session-management.py) | [📖](https://docs.praison.ai/features/sessions) |
-| ✅ Human Approval | [Example](examples/python/general/human_approval_example.py) | [📖](https://docs.praison.ai/features/approval) |
-| 🔄 Stateful Agents | [Example](examples/python/stateful/workflow-state-example.py) | [📖](https://docs.praison.ai/features/stateful-agents) |
-| 🤖 Autonomous Workflow | [Example](examples/python/general/autonomous-agent.py) | [📖](https://docs.praison.ai/features/autonomous-workflow) |
-| 📜 Rules & Instructions | [Example](#6-rules--instructions) | [📖](https://docs.praison.ai/features/rules) |
-| 🪝 Hooks | [Example](#9-hooks) | [📖](https://docs.praison.ai/features/hooks) |
-| 📈 Telemetry | [Example](examples/python/telemetry/production-telemetry-example.py) | [📖](https://docs.praison.ai/features/telemetry) |
-| 📹 Camera Integration | [Example](examples/python/camera/) | [📖](https://docs.praison.ai/features/camera-integration) |
-| 📄 Project Docs (.praison/docs/) | [Example](#docs-cli) | [📖](https://docs.praison.ai/cli/docs) |
-| 🔌 MCP Config Management | [Example](#mcp-config-cli) | [📖](https://docs.praison.ai/cli/mcp) |
-| 💬 AI Commit Messages | [Example](#ai-commit-cli) | [📖](https://docs.praison.ai/cli/commit) |
-| @ @Mentions in Prompts | [Example](#mentions-in-prompts) | [📖](https://docs.praison.ai/cli/mentions) |
-| 💾 Auto-Save Sessions | [Example](#session-management-python) | [📖](https://docs.praison.ai/cli/session) |
-| 📜 History in Context | [Example](#session-management-python) | [📖](https://docs.praison.ai/cli/session) |
+| **💻 Code & Development** | | |
+| ↳ Code Interpreter Agents | [Example](examples/python/agents/code-agent.py) | [📖](https://docs.praison.ai/features/codeagent) |
+| ↳ AI Code Editing Tools | [Example](examples/python/code/code_editing_example.py) | [📖](https://docs.praison.ai/code/editing) |
+| ↳ External Agents (All) | [Example](examples/python/code/external_agents_example.py) | [📖](https://docs.praison.ai/code/external-agents) |
+| ↳ Claude Code CLI | [Example](examples/python/code/claude_code_example.py) | [📖](https://docs.praison.ai/code/claude-code) |
+| ↳ Gemini CLI | [Example](examples/python/code/gemini_cli_example.py) | [📖](https://docs.praison.ai/code/gemini-cli) |
+| ↳ Codex CLI | [Example](examples/python/code/codex_cli_example.py) | [📖](https://docs.praison.ai/code/codex-cli) |
+| ↳ Cursor CLI | [Example](examples/python/code/cursor_cli_example.py) | [📖](https://docs.praison.ai/code/cursor-cli) |
+| **🧠 Memory & Knowledge** | | |
+| ↳ Memory (Short & Long Term) | [Example](examples/python/general/memory_example.py) | [📖](https://docs.praison.ai/concepts/memory) |
+| ↳ File-Based Memory | [Example](examples/python/general/memory_example.py) | [📖](https://docs.praison.ai/concepts/memory) |
+| ↳ Claude Memory Tool | [Example](#claude-memory-tool-cli) | [📖](https://docs.praison.ai/features/claude-memory-tool) |
+| ↳ Add Custom Knowledge | [Example](examples/python/concepts/knowledge-agents.py) | [📖](https://docs.praison.ai/features/knowledge) |
+| ↳ RAG Agents | [Example](examples/python/concepts/rag-agents.py) | [📖](https://docs.praison.ai/features/rag) |
+| ↳ Chat with PDF Agents | [Example](examples/python/concepts/chat-with-pdf.py) | [📖](https://docs.praison.ai/features/chat-with-pdf) |
+| **🔬 Research & Intelligence** | | |
+| ↳ Deep Research Agents | [Example](examples/python/agents/research-agent.py) | [📖](https://docs.praison.ai/agents/deep-research) |
+| ↳ Query Rewriter Agent | [Example](#5-query-rewriter-agent) | [📖](https://docs.praison.ai/agents/query-rewriter) |
+| ↳ Native Web Search | [Example](examples/python/agents/websearch-agent.py) | [📖](https://docs.praison.ai/agents/websearch) |
+| ↳ Built-in Search Tools | [Example](examples/python/agents/websearch-agent.py) | [📖](https://docs.praison.ai/tools/tavily) |
+| ↳ Unified Web Search | [Example](src/praisonai-agents/examples/web_search_example.py) | [📖](https://docs.praison.ai/tools/web-search) |
+| ↳ Web Fetch (Anthropic) | [Example](#web-search-web-fetch--prompt-caching) | [📖](https://docs.praison.ai/features/model-capabilities) |
+| **📋 Planning & Execution** | | |
+| ↳ Planning Mode | [Example](examples/python/agents/planning-agent.py) | [📖](https://docs.praison.ai/features/planning-mode) |
+| ↳ Planning Tools | [Example](#3-agent-with-planning-mode) | [📖](https://docs.praison.ai/features/planning-mode) |
+| ↳ Planning Reasoning | [Example](#3-agent-with-planning-mode) | [📖](https://docs.praison.ai/features/planning-mode) |
+| ↳ Prompt Chaining | [Example](examples/python/general/prompt_chaining.py) | [📖](https://docs.praison.ai/features/promptchaining) |
+| ↳ Evaluator Optimiser | [Example](examples/python/general/evaluator-optimiser.py) | [📖](https://docs.praison.ai/features/evaluator-optimiser) |
+| ↳ Orchestrator Workers | [Example](examples/python/general/orchestrator-workers.py) | [📖](https://docs.praison.ai/features/orchestrator-worker) |
+| **👥 Specialized Agents** | | |
+| ↳ Data Analyst Agent | [Example](examples/python/agents/data-analyst-agent.py) | [📖](https://docs.praison.ai/agents/data-analyst) |
+| ↳ Finance Agent | [Example](examples/python/agents/finance-agent.py) | [📖](https://docs.praison.ai/agents/finance) |
+| ↳ Shopping Agent | [Example](examples/python/agents/shopping-agent.py) | [📖](https://docs.praison.ai/agents/shopping) |
+| ↳ Recommendation Agent | [Example](examples/python/agents/recommendation-agent.py) | [📖](https://docs.praison.ai/agents/recommendation) |
+| ↳ Wikipedia Agent | [Example](examples/python/agents/wikipedia-agent.py) | [📖](https://docs.praison.ai/agents/wikipedia) |
+| ↳ Programming Agent | [Example](examples/python/agents/programming-agent.py) | [📖](https://docs.praison.ai/agents/programming) |
+| ↳ Math Agents | [Example](examples/python/agents/math-agent.py) | [📖](https://docs.praison.ai/features/mathagent) |
+| ↳ Markdown Agent | [Example](examples/python/agents/markdown-agent.py) | [📖](https://docs.praison.ai/agents/markdown) |
+| ↳ Prompt Expander Agent | [Example](#prompt-expansion) | [📖](https://docs.praison.ai/agents/prompt-expander) |
+| **🎨 Media & Multimodal** | | |
+| ↳ Image Generation Agent | [Example](examples/python/image/image-agent.py) | [📖](https://docs.praison.ai/features/image-generation) |
+| ↳ Image to Text Agent | [Example](examples/python/agents/image-to-text-agent.py) | [📖](https://docs.praison.ai/agents/image-to-text) |
+| ↳ Video Agent | [Example](examples/python/agents/video-agent.py) | [📖](https://docs.praison.ai/agents/video) |
+| ↳ Camera Integration | [Example](examples/python/camera/) | [📖](https://docs.praison.ai/features/camera-integration) |
+| **🔌 Protocols & Integration** | | |
+| ↳ MCP Transports | [Example](examples/python/mcp/mcp-transports-overview.py) | [📖](https://docs.praison.ai/mcp/transports) |
+| ↳ WebSocket MCP | [Example](examples/python/mcp/websocket-mcp.py) | [📖](https://docs.praison.ai/mcp/sse-transport) |
+| ↳ MCP Security | [Example](examples/python/mcp/mcp-security.py) | [📖](https://docs.praison.ai/mcp/transports) |
+| ↳ MCP Resumability | [Example](examples/python/mcp/mcp-resumability.py) | [📖](https://docs.praison.ai/mcp/sse-transport) |
+| ↳ MCP Config Management | [Example](#mcp-config-cli) | [📖](https://docs.praison.ai/docs/cli/mcp) |
+| ↳ LangChain Integrated Agents | [Example](examples/python/general/langchain_example.py) | [📖](https://docs.praison.ai/features/langchain) |
+| **🛡️ Safety & Control** | | |
+| ↳ Guardrails | [Example](examples/python/guardrails/comprehensive-guardrails-example.py) | [📖](https://docs.praison.ai/features/guardrails) |
+| ↳ Human Approval | [Example](examples/python/general/human_approval_example.py) | [📖](https://docs.praison.ai/features/approval) |
+| ↳ Rules & Instructions | [Example](#6-rules--instructions) | [📖](https://docs.praison.ai/features/rules) |
+| **⚙️ Advanced Features** | | |
+| ↳ Async & Parallel Processing | [Example](examples/python/general/async_example.py) | [📖](https://docs.praison.ai/features/async) |
+| ↳ Parallelisation | [Example](examples/python/general/parallelisation.py) | [📖](https://docs.praison.ai/features/parallelisation) |
+| ↳ Repetitive Agents | [Example](examples/python/concepts/repetitive-agents.py) | [📖](https://docs.praison.ai/features/repetitive) |
+| ↳ Agent Handoffs | [Example](examples/python/handoff/handoff_basic.py) | [📖](https://docs.praison.ai/features/handoffs) |
+| ↳ Stateful Agents | [Example](examples/python/stateful/workflow-state-example.py) | [📖](https://docs.praison.ai/features/stateful-agents) |
+| ↳ Autonomous Workflow | [Example](examples/python/general/autonomous-agent.py) | [📖](https://docs.praison.ai/features/autonomous-workflow) |
+| ↳ Structured Output Agents | [Example](examples/python/general/structured_agents_example.py) | [📖](https://docs.praison.ai/features/structured) |
+| ↳ Model Router | [Example](examples/python/agents/router-agent-cost-optimization.py) | [📖](https://docs.praison.ai/features/model-router) |
+| ↳ Prompt Caching | [Example](#web-search-web-fetch--prompt-caching) | [📖](https://docs.praison.ai/features/model-capabilities) |
+| ↳ Fast Context | [Example](examples/context/00_agent_fast_context_basic.py) | [📖](https://docs.praison.ai/features/fast-context) |
+| **🛠️ Tools & Configuration** | | |
+| ↳ 100+ Custom Tools | [Example](examples/python/general/tools_example.py) | [📖](https://docs.praison.ai/tools/tools) |
+| ↳ YAML Configuration | [Example](examples/cookbooks/yaml/secondary_market_research_agents.yaml) | [📖](https://docs.praison.ai/developers/agents-playbook) |
+| ↳ 100+ LLM Support | [Example](examples/python/providers/openai/openai_gpt4_example.py) | [📖](https://docs.praison.ai/models) |
+| ↳ Callback Agents | [Example](examples/python/general/advanced-callback-systems.py) | [📖](https://docs.praison.ai/features/callbacks) |
+| ↳ Hooks | [Example](#9-hooks) | [📖](https://docs.praison.ai/features/hooks) |
+| **📊 Monitoring & Management** | | |
+| ↳ Sessions Management | [Example](examples/python/sessions/comprehensive-session-management.py) | [📖](https://docs.praison.ai/features/sessions) |
+| ↳ Auto-Save Sessions | [Example](#session-management-python) | [📖](https://docs.praison.ai/docs/cli/session) |
+| ↳ History in Context | [Example](#session-management-python) | [📖](https://docs.praison.ai/docs/cli/session) |
+| ↳ Telemetry | [Example](examples/python/telemetry/production-telemetry-example.py) | [📖](https://docs.praison.ai/features/telemetry) |
+| ↳ Project Docs (.praison/docs/) | [Example](#docs-cli) | [📖](https://docs.praison.ai/docs/cli/docs) |
+| ↳ AI Commit Messages | [Example](#ai-commit-cli) | [📖](https://docs.praison.ai/docs/cli/commit) |
+| ↳ @Mentions in Prompts | [Example](#mentions-in-prompts) | [📖](https://docs.praison.ai/docs/cli/mentions) |
+| **🖥️ CLI Features** | | |
+| ↳ Slash Commands | [Example](examples/python/cli/slash_commands_example.py) | [📖](https://docs.praison.ai/docs/cli/slash-commands) |
+| ↳ Autonomy Modes | [Example](examples/python/cli/autonomy_modes_example.py) | [📖](https://docs.praison.ai/docs/cli/autonomy-modes) |
+| ↳ Cost Tracking | [Example](examples/python/cli/cost_tracking_example.py) | [📖](https://docs.praison.ai/docs/cli/cost-tracking) |
+| ↳ Repository Map | [Example](examples/python/cli/repo_map_example.py) | [📖](https://docs.praison.ai/docs/cli/repo-map) |
+| ↳ Interactive TUI | [Example](examples/python/cli/interactive_tui_example.py) | [📖](https://docs.praison.ai/docs/cli/interactive-tui) |
+| ↳ Git Integration | [Example](examples/python/cli/git_integration_example.py) | [📖](https://docs.praison.ai/docs/cli/git-integration) |
+| ↳ Sandbox Execution | [Example](examples/python/cli/sandbox_execution_example.py) | [📖](https://docs.praison.ai/docs/cli/sandbox-execution) |
+| ↳ CLI Compare | [Example](examples/compare/cli_compare_basic.py) | [📖](https://docs.praison.ai/docs/cli/compare) |
+| **🧪 Evaluation** | | |
+| ↳ Accuracy Evaluation | [Example](examples/eval/accuracy_example.py) | [📖](https://docs.praison.ai/docs/cli/eval) |
+| ↳ Performance Evaluation | [Example](examples/eval/performance_example.py) | [📖](https://docs.praison.ai/docs/cli/eval) |
+| ↳ Reliability Evaluation | [Example](examples/eval/reliability_example.py) | [📖](https://docs.praison.ai/docs/cli/eval) |
+| ↳ Criteria Evaluation | [Example](examples/eval/criteria_example.py) | [📖](https://docs.praison.ai/docs/cli/eval) |
+| **🎯 Agent Skills** | | |
+| ↳ Skills Management | [Example](examples/skills/basic_skill_usage.py) | [📖](https://docs.praison.ai/features/skills) |
+| ↳ Custom Skills | [Example](examples/skills/custom_skill_example.py) | [📖](https://docs.praison.ai/features/skills) |
+| **⏰ 24/7 Scheduling** | | |
+| ↳ Agent Scheduler | [Example](examples/python/scheduled_agents/news_checker_live.py) | [📖](https://docs.praison.ai/docs/cli/scheduler) |

 ---

@@ -954,6 +994,32 @@ export OPENAI_API_KEY=xxxxxxxxxxxxxxxxxxxxxx
 praisonai --auto create a movie script about Robots in Mars
 ```

+### Interactive Mode CLI:
+```bash
+# Start interactive terminal mode (inspired by Gemini CLI, Codex CLI, Claude Code)
+praisonai --interactive
+praisonai -i
+
+# Features:
+# - Streaming responses (no boxes)
+# - Built-in tools: read_file, write_file, list_files, execute_command, internet_search
+# - Slash commands: /help, /exit, /tools, /clear
+
+# Chat mode - single prompt with interactive style (for testing/scripting)
+# Use --chat (or --chat-mode for backward compatibility)
+praisonai "list files in current folder" --chat
+praisonai "search the web for AI news" --chat
+praisonai "read README.md" --chat
+```
+
+### Chat UI (Web Interface):
+```bash
+# Start web-based Chainlit chat interface (requires praisonai[chat])
+pip install "praisonai[chat]"
+praisonai chat
+# Opens browser at http://localhost:8084
+```
+
 ### Query Rewriting (works with any command):
 ```bash
 # Rewrite query for better results (uses QueryRewriterAgent)
@@ -1014,6 +1080,20 @@ praisonai "Complex analysis task" --planning --planning-reasoning
 praisonai "Task" --planning --auto-approve-plan
 ```

+### Tool Approval CLI:
+```bash
+# Auto-approve ALL tool executions (use with caution!)
+praisonai "run ls command" --trust
+
+# Auto-approve tools up to a risk level (prompt for higher)
+# Levels: low, medium, high, critical
+praisonai "write to file" --approve-level high  # Prompts for critical tools only
+praisonai "task" --approve-level medium         # Prompts for high and critical
+
+# Default behavior (no flags): prompts for all dangerous tools
+praisonai "run shell command"  # Will prompt for approval
+```
+
 ### Memory CLI:
 ```bash
 # Enable memory for agent (persists across sessions)
@@ -1120,6 +1200,15 @@ praisonai workflow auto "Research AI trends" --pattern parallel
 # Generate routing workflow (classifier routes to specialists)
 praisonai workflow auto "Build a chatbot" --pattern routing

+# Generate orchestrator-workers workflow (central orchestrator delegates)
+praisonai workflow auto "Comprehensive market analysis" --pattern orchestrator-workers
+
+# Generate evaluator-optimizer workflow (iterative refinement)
+praisonai workflow auto "Write and refine article" --pattern evaluator-optimizer
+
+# Specify output file
+praisonai workflow auto "Build a chatbot" --pattern routing
+
 # Specify output file
 praisonai workflow auto "Research AI" --pattern sequential --output my_workflow.yaml
 ```
@@ -1129,7 +1218,7 @@ praisonai workflow auto "Research AI" --pattern sequential --output my_workflow.
 |------|-------------|
 | `--workflow-var key=value` | Set workflow variable (can be repeated) |
 | `--var key=value` | Set variable for YAML workflows |
-| `--pattern <pattern>` | Pattern for auto-generation (sequential, parallel, routing, loop) |
+| `--pattern <pattern>` | Pattern for auto-generation (sequential, parallel, routing, loop, orchestrator-workers, evaluator-optimizer) |
 | `--output <file>` | Output file for auto-generation |
 | `--llm <model>` | LLM model (e.g., openai/gpt-4o-mini) |
 | `--tools <tools>` | Tools (comma-separated, e.g., tavily) |
@@ -1175,6 +1264,21 @@ praisonai "Analyze this data" --metrics
 praisonai "Complex task" --metrics --planning
 ```

+### Scheduler CLI:
+
+```bash
+praisonai schedule start <name> "task" --interval hourly
+praisonai schedule list
+praisonai schedule logs <name> [--follow]
+praisonai schedule stop <name>
+praisonai schedule restart <name>
+praisonai schedule delete <name>
+praisonai schedule describe <name>
+praisonai schedule save <name> [file.yaml]
+praisonai schedule "task" --interval hourly  # foreground mode
+praisonai schedule agents.yaml  # foreground mode
+```
+
 ### Image Processing CLI:
 ```bash
 # Process images with vision-based tasks
@@ -1361,8 +1465,13 @@ praisonai "Complex analysis" --router --router-provider anthropic
 # Router automatically selects:
 # - Simple tasks → gpt-4o-mini, claude-3-haiku
 # - Complex tasks → gpt-4-turbo, claude-3-opus
+
+# Create workflow with model routing template
+praisonai workflow create --template model-routing --output my_workflow.yaml
 ```

+Custom models can be configured in `agents.yaml`. See [Model Router Docs](https://docs.praison.ai/features/model-router) for details.
+
 ### Flow Display CLI:
 ```bash
 # Enable visual workflow tracking
@@ -1414,13 +1523,31 @@ praisonai mcp help

 ### AI Commit CLI:
 ```bash
-# Generate AI commit message for staged changes
+# Full auto mode: stage all, security check, commit, and push
+praisonai commit -a
+
+# Interactive mode (requires git add first)
 praisonai commit

-# Generate, commit, and push
+# Interactive with auto-push
 praisonai commit --push
+
+# Skip security check (not recommended)
+praisonai commit -a --no-verify
 ```

+**Features:**
+- 🤖 AI-generated conventional commit messages
+- 🔒 Built-in security scanning (API keys, passwords, secrets, sensitive files)
+- 📦 Auto-staging with `-a` flag
+- 🚀 Auto-push in full auto mode
+- ✏️ Edit message before commit in interactive mode
+
+**Security Detection:**
+- API keys, secrets, tokens (AWS, GitHub, GitLab, Slack)
+- Passwords and private keys
+- Sensitive files (`.env`, `id_rsa`, `.pem`, `.key`, etc.)
+
 ### Serve CLI (API Server):
 ```bash
 # Start API server for agents defined in YAML
@@ -1451,6 +1578,55 @@ export N8N_API_KEY="your-api-key"
 praisonai agents.yaml --n8n
 ```

+### External Agents CLI:
+
+Use external AI coding CLI tools (Claude Code, Gemini CLI, Codex CLI, Cursor CLI) as agent tools:
+
+```bash
+# Use Claude Code for coding tasks
+praisonai "Refactor the auth module" --external-agent claude
+
+# Use Gemini CLI for code analysis
+praisonai "Analyze codebase architecture" --external-agent gemini
+
+# Use OpenAI Codex CLI
+praisonai "Fix all bugs in src/" --external-agent codex
+
+# Use Cursor CLI
+praisonai "Add comprehensive tests" --external-agent cursor
+```
+
+**Python API:**
+```python
+from praisonai.integrations import (
+    ClaudeCodeIntegration,
+    GeminiCLIIntegration,
+    CodexCLIIntegration,
+    CursorCLIIntegration
+)
+
+# Create integration
+claude = ClaudeCodeIntegration(workspace="/project")
+
+# Execute a coding task
+result = await claude.execute("Refactor the auth module")
+
+# Use as agent tool
+from praisonai import Agent
+tool = claude.as_tool()
+agent = Agent(tools=[tool])
+```
+
+**Environment Variables:**
+```bash
+export ANTHROPIC_API_KEY=your-key  # Claude Code
+export GEMINI_API_KEY=your-key     # Gemini CLI
+export OPENAI_API_KEY=your-key     # Codex CLI
+export CURSOR_API_KEY=your-key     # Cursor CLI
+```
+
+See [External Agents Documentation](https://docs.praison.ai/code/external-agents) for more details.
+
 ### @Mentions in Prompts:
 ```bash
 # Include file content in prompt
@@ -1676,32 +1852,32 @@ app.include_router(a2a.get_router())

 | Feature | Docs |
 |---------|:----:|
-| 🔄 Query Rewrite - RAG optimization | [📖](https://docs.praison.ai/cli/query-rewrite) |
-| 🔬 Deep Research - Automated research | [📖](https://docs.praison.ai/cli/deep-research) |
-| 📋 Planning - Step-by-step execution | [📖](https://docs.praison.ai/cli/planning) |
-| 💾 Memory - Persistent agent memory | [📖](https://docs.praison.ai/cli/memory) |
-| 📜 Rules - Auto-discovered instructions | [📖](https://docs.praison.ai/cli/rules) |
-| 🔄 Workflow - Multi-step workflows | [📖](https://docs.praison.ai/cli/workflow) |
-| 🪝 Hooks - Event-driven actions | [📖](https://docs.praison.ai/cli/hooks) |
-| 🧠 Claude Memory - Anthropic memory tool | [📖](https://docs.praison.ai/cli/claude-memory) |
-| 🛡️ Guardrail - Output validation | [📖](https://docs.praison.ai/cli/guardrail) |
-| 📊 Metrics - Token usage tracking | [📖](https://docs.praison.ai/cli/metrics) |
-| 🖼️ Image - Vision processing | [📖](https://docs.praison.ai/cli/image) |
-| 📡 Telemetry - Usage monitoring | [📖](https://docs.praison.ai/cli/telemetry) |
-| 🔌 MCP - Model Context Protocol | [📖](https://docs.praison.ai/cli/mcp) |
-| ⚡ Fast Context - Codebase search | [📖](https://docs.praison.ai/cli/fast-context) |
-| 📚 Knowledge - RAG management | [📖](https://docs.praison.ai/cli/knowledge) |
-| 💬 Session - Conversation management | [📖](https://docs.praison.ai/cli/session) |
-| 🔧 Tools - Tool discovery | [📖](https://docs.praison.ai/cli/tools) |
-| 🤝 Handoff - Agent delegation | [📖](https://docs.praison.ai/cli/handoff) |
-| 🧠 Auto Memory - Memory extraction | [📖](https://docs.praison.ai/cli/auto-memory) |
-| 📋 Todo - Task management | [📖](https://docs.praison.ai/cli/todo) |
-| 🎯 Router - Smart model selection | [📖](https://docs.praison.ai/cli/router) |
-| 📈 Flow Display - Visual workflow | [📖](https://docs.praison.ai/cli/flow-display) |
-| ✨ Prompt Expansion - Detailed prompts | [📖](https://docs.praison.ai/cli/prompt-expansion) |
-| 🌐 Web Search - Real-time search | [📖](https://docs.praison.ai/cli/web-search) |
-| 📥 Web Fetch - URL content retrieval | [📖](https://docs.praison.ai/cli/web-fetch) |
-| 💾 Prompt Caching - Cost reduction | [📖](https://docs.praison.ai/cli/prompt-caching) |
+| 🔄 Query Rewrite - RAG optimization | [📖](https://docs.praison.ai/docs/cli/query-rewrite) |
+| 🔬 Deep Research - Automated research | [📖](https://docs.praison.ai/docs/cli/deep-research) |
+| 📋 Planning - Step-by-step execution | [📖](https://docs.praison.ai/docs/cli/planning) |
+| 💾 Memory - Persistent agent memory | [📖](https://docs.praison.ai/docs/cli/memory) |
+| 📜 Rules - Auto-discovered instructions | [📖](https://docs.praison.ai/docs/cli/rules) |
+| 🔄 Workflow - Multi-step workflows | [📖](https://docs.praison.ai/docs/cli/workflow) |
+| 🪝 Hooks - Event-driven actions | [📖](https://docs.praison.ai/docs/cli/hooks) |
+| 🧠 Claude Memory - Anthropic memory tool | [📖](https://docs.praison.ai/docs/cli/claude-memory) |
+| 🛡️ Guardrail - Output validation | [📖](https://docs.praison.ai/docs/cli/guardrail) |
+| 📊 Metrics - Token usage tracking | [📖](https://docs.praison.ai/docs/cli/metrics) |
+| 🖼️ Image - Vision processing | [📖](https://docs.praison.ai/docs/cli/image) |
+| 📡 Telemetry - Usage monitoring | [📖](https://docs.praison.ai/docs/cli/telemetry) |
+| 🔌 MCP - Model Context Protocol | [📖](https://docs.praison.ai/docs/cli/mcp) |
+| ⚡ Fast Context - Codebase search | [📖](https://docs.praison.ai/docs/cli/fast-context) |
+| 📚 Knowledge - RAG management | [📖](https://docs.praison.ai/docs/cli/knowledge) |
+| 💬 Session - Conversation management | [📖](https://docs.praison.ai/docs/cli/session) |
+| 🔧 Tools - Tool discovery | [📖](https://docs.praison.ai/docs/cli/tools) |
+| 🤝 Handoff - Agent delegation | [📖](https://docs.praison.ai/docs/cli/handoff) |
+| 🧠 Auto Memory - Memory extraction | [📖](https://docs.praison.ai/docs/cli/auto-memory) |
+| 📋 Todo - Task management | [📖](https://docs.praison.ai/docs/cli/todo) |
+| 🎯 Router - Smart model selection | [📖](https://docs.praison.ai/docs/cli/router) |
+| 📈 Flow Display - Visual workflow | [📖](https://docs.praison.ai/docs/cli/flow-display) |
+| ✨ Prompt Expansion - Detailed prompts | [📖](https://docs.praison.ai/docs/cli/prompt-expansion) |
+| 🌐 Web Search - Real-time search | [📖](https://docs.praison.ai/docs/cli/web-search) |
+| 📥 Web Fetch - URL content retrieval | [📖](https://docs.praison.ai/docs/cli/web-fetch) |
+| 💾 Prompt Caching - Cost reduction | [📖](https://docs.praison.ai/docs/cli/prompt-caching) |

 ---

@@ -2253,7 +2429,7 @@ PraisonAI provides zero-dependency persistent memory for agents. For detailed ex
 - 📖 **Wikipedia Agent** - Search and extract Wikipedia information
 - 💻 **Programming Agent** - Code development and analysis
 - 📝 **Markdown Agent** - Generate and format Markdown content
- 🔀 **Router Agent** - Dynamic task routing with cost optimisation
+- 🔀 **Model Router** - Smart model selection based on task complexity

 ### MCP Protocol

@@ -2274,6 +2450,7 @@ PraisonAI provides zero-dependency persistent memory for agents. For detailed ex
 - 🤝 **Agent Handoffs** - Transfer context between specialised agents
 - 🛡️ **Guardrails** - Input/output validation and safety checks
 - ✅ **Human Approval** - Require human confirmation for critical actions
+- 🔐 **Tool Approval CLI** - `--trust` (auto-approve all) and `--approve-level` (risk-based approval)
 - 💬 **Sessions Management** - Isolated conversation contexts
 - 🔄 **Stateful Agents** - Maintain state across interactions

@@ -0,0 +1,33 @@
+# PraisonAI CLI - Benchmark Results
+
+**Generated:** 2025-12-19 16:04:40
+
+## Import Times
+
+| Module | Time (ms) | Target | Status |
+|--------|-----------|--------|--------|
+| praisonai | 4.11 | <10 | ✅ |
+| message_queue | 6.87 | <50 | ✅ |
+| at_mentions | 0.58 | <50 | ✅ |
+| profiler | 0.90 | <10 | ✅ |
+
+## Instantiation Times
+
+| Class | Time (μs) | Target | Status |
+|-------|-----------|--------|--------|
+| MessageQueue | 0.26 | <100 | ✅ |
+| StateManager | 0.30 | <100 | ✅ |
+| FileSearchService | 8.42 | <100 | ✅ |
+
+## Lazy Loading
+
+| Operation | Time (ms) |
+|-----------|----------|
+| PraisonAI lazy load | 3886.47 |
+
+## How to Reproduce
+
+```bash
+cd praisonai
+python benchmarks/simple_benchmark.py
+```
@@ -0,0 +1,231 @@
+#!/usr/bin/env python3
+"""
+PraisonAI CLI - Performance Benchmark
+
+Measures import times and module loading performance.
+
+Usage:
+    python benchmarks/simple_benchmark.py
+    python benchmarks/simple_benchmark.py --no-save
+"""
+
+import time
+import sys
+import os
+import argparse
+from datetime import datetime
+
+
+ITERATIONS = 100
+
+
+def measure_import(module_name: str) -> float:
+    """Measure import time in milliseconds."""
+    # Clear from cache if present
+    modules_to_remove = [k for k in sys.modules.keys() if k.startswith(module_name)]
+    for mod in modules_to_remove:
+        del sys.modules[mod]
+    
+    start = time.perf_counter()
+    __import__(module_name)
+    return (time.perf_counter() - start) * 1000
+
+
+def measure_instantiation(create_fn, iterations=ITERATIONS) -> float:
+    """Measure average instantiation time in microseconds."""
+    times = []
+    for _ in range(iterations):
+        start = time.perf_counter()
+        create_fn()
+        times.append((time.perf_counter() - start) * 1_000_000)
+    return sum(times) / len(times)
+
+
+def run_benchmark():
+    """Run the benchmark."""
+    results = {}
+    
+    print('=' * 60)
+    print('PraisonAI CLI - Performance Benchmark')
+    print('=' * 60)
+    print(f'\nIterations: {ITERATIONS}')
+    print('Metrics: Import time (ms), Instantiation time (μs)\n')
+    
+    # 1. Package import time
+    print("1. Testing package imports...")
+    
+    # Clear cache
+    modules_to_remove = [k for k in sys.modules.keys() if 'praisonai' in k]
+    for mod in modules_to_remove:
+        del sys.modules[mod]
+    
+    start = time.perf_counter()
+    import praisonai
+    results['praisonai_import_ms'] = (time.perf_counter() - start) * 1000
+    print(f"   import praisonai: {results['praisonai_import_ms']:.2f}ms")
+    
+    # 2. CLI features import
+    print("\n2. Testing CLI features imports...")
+    
+    start = time.perf_counter()
+    from praisonai.cli.features.message_queue import MessageQueue
+    results['message_queue_import_ms'] = (time.perf_counter() - start) * 1000
+    print(f"   message_queue: {results['message_queue_import_ms']:.2f}ms")
+    
+    start = time.perf_counter()
+    from praisonai.cli.features.at_mentions import CombinedCompleter
+    results['at_mentions_import_ms'] = (time.perf_counter() - start) * 1000
+    print(f"   at_mentions: {results['at_mentions_import_ms']:.2f}ms")
+    
+    # 3. Profiler import (should be fast, standalone)
+    print("\n3. Testing profiler import...")
+    start = time.perf_counter()
+    from praisonai.profiler import Profiler
+    results['profiler_import_ms'] = (time.perf_counter() - start) * 1000
+    print(f"   profiler: {results['profiler_import_ms']:.2f}ms")
+    
+    # 4. Instantiation benchmarks
+    print(f"\n4. Testing instantiation ({ITERATIONS} iterations)...")
+    
+    from praisonai.cli.features.message_queue import MessageQueue, StateManager
+    from praisonai.cli.features.at_mentions import FileSearchService
+    
+    results['MessageQueue_us'] = measure_instantiation(lambda: MessageQueue())
+    print(f"   MessageQueue: {results['MessageQueue_us']:.2f}μs")
+    
+    results['StateManager_us'] = measure_instantiation(lambda: StateManager())
+    print(f"   StateManager: {results['StateManager_us']:.2f}μs")
+    
+    results['FileSearchService_us'] = measure_instantiation(
+        lambda: FileSearchService(root_dir=".")
+    )
+    print(f"   FileSearchService: {results['FileSearchService_us']:.2f}μs")
+    
+    # 5. PraisonAI class (lazy load)
+    print("\n5. Testing PraisonAI class access (lazy load)...")
+    start = time.perf_counter()
+    PraisonAI = praisonai.PraisonAI
+    results['PraisonAI_lazy_ms'] = (time.perf_counter() - start) * 1000
+    print(f"   PraisonAI lazy load: {results['PraisonAI_lazy_ms']:.2f}ms")
+    
+    return results
+
+
+def print_summary(results: dict):
+    """Print benchmark summary."""
+    print('\n' + '=' * 60)
+    print('SUMMARY')
+    print('=' * 60)
+    
+    print('\nImport Times (target: <50ms for features, <10ms for package):')
+    print(f"  {'Module':<25} {'Time':>10} {'Status':>10}")
+    print(f"  {'-'*25} {'-'*10} {'-'*10}")
+    
+    import_checks = [
+        ('praisonai', results['praisonai_import_ms'], 10),
+        ('message_queue', results['message_queue_import_ms'], 50),
+        ('at_mentions', results['at_mentions_import_ms'], 50),
+        ('profiler', results['profiler_import_ms'], 10),
+    ]
+    
+    all_pass = True
+    for name, time_ms, threshold in import_checks:
+        status = '✅ PASS' if time_ms < threshold else '❌ SLOW'
+        if time_ms >= threshold:
+            all_pass = False
+        print(f"  {name:<25} {time_ms:>8.2f}ms {status:>10}")
+    
+    print('\nInstantiation Times (target: <100μs):')
+    print(f"  {'Class':<25} {'Time':>10} {'Status':>10}")
+    print(f"  {'-'*25} {'-'*10} {'-'*10}")
+    
+    inst_checks = [
+        ('MessageQueue', results['MessageQueue_us'], 100),
+        ('StateManager', results['StateManager_us'], 100),
+        ('FileSearchService', results['FileSearchService_us'], 100),
+    ]
+    
+    for name, time_us, threshold in inst_checks:
+        status = '✅ PASS' if time_us < threshold else '❌ SLOW'
+        if time_us >= threshold:
+            all_pass = False
+        print(f"  {name:<25} {time_us:>8.2f}μs {status:>10}")
+    
+    print('\n' + '=' * 60)
+    if all_pass:
+        print('✅ All benchmarks PASSED')
+    else:
+        print('❌ Some benchmarks FAILED')
+    print('=' * 60)
+    
+    return all_pass
+
+
+def save_results(results: dict):
+    """Save benchmark results to markdown file."""
+    filepath = os.path.join(os.path.dirname(__file__), 'BENCHMARK_RESULTS.md')
+    
+    with open(filepath, 'w') as f:
+        f.write('# PraisonAI CLI - Benchmark Results\n\n')
+        f.write(f'**Generated:** {datetime.now().strftime("%Y-%m-%d %H:%M:%S")}\n\n')
+        
+        f.write('## Import Times\n\n')
+        f.write('| Module | Time (ms) | Target | Status |\n')
+        f.write('|--------|-----------|--------|--------|\n')
+        
+        import_data = [
+            ('praisonai', results['praisonai_import_ms'], 10),
+            ('message_queue', results['message_queue_import_ms'], 50),
+            ('at_mentions', results['at_mentions_import_ms'], 50),
+            ('profiler', results['profiler_import_ms'], 10),
+        ]
+        
+        for name, time_ms, threshold in import_data:
+            status = '✅' if time_ms < threshold else '❌'
+            f.write(f'| {name} | {time_ms:.2f} | <{threshold} | {status} |\n')
+        
+        f.write('\n## Instantiation Times\n\n')
+        f.write('| Class | Time (μs) | Target | Status |\n')
+        f.write('|-------|-----------|--------|--------|\n')
+        
+        inst_data = [
+            ('MessageQueue', results['MessageQueue_us'], 100),
+            ('StateManager', results['StateManager_us'], 100),
+            ('FileSearchService', results['FileSearchService_us'], 100),
+        ]
+        
+        for name, time_us, threshold in inst_data:
+            status = '✅' if time_us < threshold else '❌'
+            f.write(f'| {name} | {time_us:.2f} | <{threshold} | {status} |\n')
+        
+        f.write('\n## Lazy Loading\n\n')
+        f.write(f'| Operation | Time (ms) |\n')
+        f.write(f'|-----------|----------|\n')
+        f.write(f'| PraisonAI lazy load | {results["PraisonAI_lazy_ms"]:.2f} |\n')
+        
+        f.write('\n## How to Reproduce\n\n')
+        f.write('```bash\n')
+        f.write('cd praisonai\n')
+        f.write('python benchmarks/simple_benchmark.py\n')
+        f.write('```\n')
+    
+    print(f'\nResults saved to: {filepath}')
+    return filepath
+
+
+def main():
+    parser = argparse.ArgumentParser(description='PraisonAI CLI Performance Benchmark')
+    parser.add_argument('--no-save', action='store_true', help='Do not save results to file')
+    args = parser.parse_args()
+    
+    results = run_benchmark()
+    all_pass = print_summary(results)
+    
+    if not args.no_save:
+        save_results(results)
+    
+    return 0 if all_pass else 1
+
+
+if __name__ == '__main__':
+    sys.exit(main())
@@ -3,8 +3,8 @@ class Praisonai < Formula
  
    desc "AI tools for various AI applications"
    homepage "https://github.com/MervinPraison/PraisonAI"
-    url "https://github.com/MervinPraison/PraisonAI/archive/refs/tags/v2.3.41.tar.gz"
-    sha256 `curl -sL https://github.com/MervinPraison/PraisonAI/archive/refs/tags/v2.3.41.tar.gz | shasum -a 256`.split.first
+    url "https://github.com/MervinPraison/PraisonAI/archive/refs/tags/v2.3.75.tar.gz"
+    sha256 `curl -sL https://github.com/MervinPraison/PraisonAI/archive/refs/tags/v2.3.75.tar.gz | shasum -a 256`.split.first
    license "MIT"
  
    depends_on "[email protected]"
@@ -1,31 +1,34 @@
+# Suppress crewai.cli.config logger BEFORE any imports to prevent INFO log
+import logging
+logging.getLogger('crewai.cli.config').setLevel(logging.ERROR)
+
 # Disable OpenTelemetry SDK
 import os
 os.environ["OTEL_SDK_DISABLED"] = "true"
 os.environ["EC_TELEMETRY"] = "false"
-from .cli import PraisonAI
+
+# Version is lightweight, import directly
 from .version import __version__

-# Re-export all classes from praisonaiagents to enable:
-# from praisonai import Agent, Task, PraisonAIAgents
-try:
-    import praisonaiagents
-    # Import all symbols from praisonaiagents using * import
-    from praisonaiagents import *
-except ImportError:
-    # If praisonaiagents is not available, these imports will fail gracefully
-    pass
-
-# Define __all__ to include both PraisonAI core classes and praisonaiagents exports
+# Define __all__ for lazy loading
 __all__ = [
-    # Core PraisonAI classes
    'PraisonAI',
    '__version__',
 ]

-# Dynamically extend __all__ with praisonaiagents exports
-try:
-    import praisonaiagents
-    __all__.extend(praisonaiagents.__all__)
-except (ImportError, AttributeError):
-    # If praisonaiagents is not available or doesn't have __all__, fail gracefully
-    pass
+# Lazy loading for heavy imports
+def __getattr__(name):
+    """Lazy load heavy modules to improve import time."""
+    if name == 'PraisonAI':
+        from .cli import PraisonAI
+        return PraisonAI
+    
+    # Try praisonaiagents exports
+    try:
+        import praisonaiagents
+        if hasattr(praisonaiagents, name):
+            return getattr(praisonaiagents, name)
+    except ImportError:
+        pass
+    
+    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
@@ -0,0 +1,337 @@
+"""
+Agent Scheduler for PraisonAI - Run agents periodically 24/7.
+
+This module provides scheduling capabilities for running PraisonAI agents
+at regular intervals, enabling 24/7 autonomous agent operations.
+
+Example:
+    # Run news checker every hour
+    from praisonai.agent_scheduler import AgentScheduler
+    from praisonai_agents import Agent
+    
+    agent = Agent(
+        name="NewsChecker",
+        instructions="Check latest AI news and summarize",
+        tools=[search_tool]
+    )
+    
+    scheduler = AgentScheduler(agent, task="Check latest AI news")
+    scheduler.start(schedule_expr="hourly")
+"""
+
+import threading
+import time
+import logging
+from datetime import datetime
+from typing import Optional, Dict, Any, Callable
+from abc import ABC, abstractmethod
+
+logger = logging.getLogger(__name__)
+
+
+class ScheduleParser:
+    """Parse schedule expressions into intervals."""
+    
+    @staticmethod
+    def parse(schedule_expr: str) -> int:
+        """
+        Parse schedule expression and return interval in seconds.
+        
+        Supported formats:
+        - "daily" -> 86400 seconds
+        - "hourly" -> 3600 seconds
+        - "*/30m" -> 1800 seconds (every 30 minutes)
+        - "*/1h" -> 3600 seconds (every 1 hour)
+        - "60" -> 60 seconds (plain number)
+        
+        Args:
+            schedule_expr: Schedule expression string
+            
+        Returns:
+            Interval in seconds
+            
+        Raises:
+            ValueError: If schedule format is not supported
+        """
+        schedule_expr = schedule_expr.strip().lower()
+        
+        if schedule_expr == "daily":
+            return 86400
+        elif schedule_expr == "hourly":
+            return 3600
+        elif schedule_expr.isdigit():
+            return int(schedule_expr)
+        elif schedule_expr.startswith("*/"):
+            interval_part = schedule_expr[2:]
+            if interval_part.endswith("m"):
+                minutes = int(interval_part[:-1])
+                return minutes * 60
+            elif interval_part.endswith("h"):
+                hours = int(interval_part[:-1])
+                return hours * 3600
+            elif interval_part.endswith("s"):
+                return int(interval_part[:-1])
+            else:
+                return int(interval_part)
+        else:
+            raise ValueError(f"Unsupported schedule format: {schedule_expr}")
+
+
+class AgentExecutorInterface(ABC):
+    """Abstract interface for agent execution."""
+    
+    @abstractmethod
+    def execute(self, task: str) -> Any:
+        """Execute the agent with given task."""
+        pass
+
+
+class PraisonAgentExecutor(AgentExecutorInterface):
+    """Executor for PraisonAI agents."""
+    
+    def __init__(self, agent):
+        """
+        Initialize executor with a PraisonAI agent.
+        
+        Args:
+            agent: PraisonAI Agent instance
+        """
+        self.agent = agent
+        
+    def execute(self, task: str) -> Any:
+        """
+        Execute the agent with given task.
+        
+        Args:
+            task: Task description for the agent
+            
+        Returns:
+            Agent execution result
+        """
+        try:
+            result = self.agent.start(task)
+            return result
+        except Exception as e:
+            logger.error(f"Agent execution failed: {e}")
+            raise
+
+
+class AgentScheduler:
+    """
+    Scheduler for running PraisonAI agents periodically.
+    
+    Features:
+    - Interval-based scheduling (hourly, daily, custom)
+    - Thread-safe operation
+    - Automatic retry on failure
+    - Execution logging and monitoring
+    - Graceful shutdown
+    
+    Example:
+        scheduler = AgentScheduler(agent, task="Check news")
+        scheduler.start(schedule_expr="hourly", max_retries=3)
+        # Agent runs every hour automatically
+        scheduler.stop()  # Stop when needed
+    """
+    
+    def __init__(
+        self,
+        agent,
+        task: str,
+        config: Optional[Dict[str, Any]] = None,
+        on_success: Optional[Callable] = None,
+        on_failure: Optional[Callable] = None
+    ):
+        """
+        Initialize agent scheduler.
+        
+        Args:
+            agent: PraisonAI Agent instance
+            task: Task description to execute
+            config: Optional configuration dict
+            on_success: Callback function on successful execution
+            on_failure: Callback function on failed execution
+        """
+        self.agent = agent
+        self.task = task
+        self.config = config or {}
+        self.on_success = on_success
+        self.on_failure = on_failure
+        
+        self.is_running = False
+        self._stop_event = threading.Event()
+        self._thread = None
+        self._executor = PraisonAgentExecutor(agent)
+        self._execution_count = 0
+        self._success_count = 0
+        self._failure_count = 0
+        
+    def start(
+        self,
+        schedule_expr: str,
+        max_retries: int = 3,
+        run_immediately: bool = False
+    ) -> bool:
+        """
+        Start scheduled agent execution.
+        
+        Args:
+            schedule_expr: Schedule expression (e.g., "hourly", "*/1h", "3600")
+            max_retries: Maximum retry attempts on failure
+            run_immediately: If True, run agent immediately before starting schedule
+            
+        Returns:
+            True if scheduler started successfully
+        """
+        if self.is_running:
+            logger.warning("Scheduler is already running")
+            return False
+            
+        try:
+            interval = ScheduleParser.parse(schedule_expr)
+            self.is_running = True
+            self._stop_event.clear()
+            
+            logger.info(f"Starting agent scheduler: {self.agent.name if hasattr(self.agent, 'name') else 'Agent'}")
+            logger.info(f"Task: {self.task}")
+            logger.info(f"Schedule: {schedule_expr} ({interval}s interval)")
+            logger.info(f"Max retries: {max_retries}")
+            
+            # Run immediately if requested
+            if run_immediately:
+                logger.info("Running agent immediately before starting schedule...")
+                self._execute_with_retry(max_retries)
+            
+            self._thread = threading.Thread(
+                target=self._run_schedule,
+                args=(interval, max_retries),
+                daemon=True
+            )
+            self._thread.start()
+            
+            logger.info("Agent scheduler started successfully")
+            return True
+            
+        except Exception as e:
+            logger.error(f"Failed to start scheduler: {e}")
+            self.is_running = False
+            return False
+    
+    def stop(self) -> bool:
+        """
+        Stop the scheduler gracefully.
+        
+        Returns:
+            True if stopped successfully
+        """
+        if not self.is_running:
+            logger.info("Scheduler is not running")
+            return True
+            
+        logger.info("Stopping agent scheduler...")
+        self._stop_event.set()
+        
+        if self._thread and self._thread.is_alive():
+            self._thread.join(timeout=10)
+            
+        self.is_running = False
+        logger.info("Agent scheduler stopped")
+        logger.info(f"Execution stats - Total: {self._execution_count}, Success: {self._success_count}, Failed: {self._failure_count}")
+        return True
+    
+    def get_stats(self) -> Dict[str, Any]:
+        """
+        Get execution statistics.
+        
+        Returns:
+            Dictionary with execution stats
+        """
+        return {
+            "is_running": self.is_running,
+            "total_executions": self._execution_count,
+            "successful_executions": self._success_count,
+            "failed_executions": self._failure_count,
+            "success_rate": (self._success_count / self._execution_count * 100) if self._execution_count > 0 else 0
+        }
+    
+    def _run_schedule(self, interval: int, max_retries: int):
+        """Internal method to run scheduled agent executions."""
+        while not self._stop_event.is_set():
+            logger.info(f"[{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}] Starting scheduled agent execution")
+            
+            self._execute_with_retry(max_retries)
+            
+            # Wait for next scheduled time
+            logger.info(f"Next execution in {interval} seconds ({interval/3600:.1f} hours)")
+            self._stop_event.wait(interval)
+    
+    def _execute_with_retry(self, max_retries: int):
+        """Execute agent with retry logic."""
+        self._execution_count += 1
+        success = False
+        
+        for attempt in range(max_retries):
+            try:
+                logger.info(f"Attempt {attempt + 1}/{max_retries}")
+                result = self._executor.execute(self.task)
+                
+                logger.info(f"Agent execution successful on attempt {attempt + 1}")
+                logger.info(f"Result: {result}")
+                
+                self._success_count += 1
+                success = True
+                
+                if self.on_success:
+                    self.on_success(result)
+                    
+                break
+                
+            except Exception as e:
+                logger.error(f"Agent execution failed on attempt {attempt + 1}: {e}")
+                
+                if attempt < max_retries - 1:
+                    wait_time = 30 * (attempt + 1)  # Exponential backoff
+                    logger.info(f"Waiting {wait_time}s before retry...")
+                    time.sleep(wait_time)
+        
+        if not success:
+            self._failure_count += 1
+            logger.error(f"Agent execution failed after {max_retries} attempts")
+            
+            if self.on_failure:
+                self.on_failure(f"Failed after {max_retries} attempts")
+    
+    def execute_once(self) -> Any:
+        """
+        Execute agent immediately (one-time execution).
+        
+        Returns:
+            Agent execution result
+        """
+        logger.info("Executing agent once")
+        try:
+            result = self._executor.execute(self.task)
+            logger.info(f"One-time execution successful: {result}")
+            return result
+        except Exception as e:
+            logger.error(f"One-time execution failed: {e}")
+            raise
+
+
+def create_agent_scheduler(
+    agent,
+    task: str,
+    config: Optional[Dict[str, Any]] = None
+) -> AgentScheduler:
+    """
+    Factory function to create agent scheduler.
+    
+    Args:
+        agent: PraisonAI Agent instance
+        task: Task description
+        config: Optional configuration
+        
+    Returns:
+        Configured AgentScheduler instance
+    """
+    return AgentScheduler(agent, task, config)
@@ -51,14 +51,140 @@ try:
 except ImportError:
    PRAISONAI_TOOLS_AVAILABLE = False

+# LiteLLM availability check for multi-provider support
+LITELLM_AVAILABLE = False
+try:
+    import litellm  # noqa: F401 - imported for availability check
+    LITELLM_AVAILABLE = True
+except ImportError:
+    pass
+
 logging.basicConfig(level=os.environ.get('LOGLEVEL', 'INFO').upper(), format='%(asctime)s - %(levelname)s - %(message)s')

-# Define Pydantic models outside of the generate method
+# =============================================================================
+# Available Tools List (shared between generators)
+# =============================================================================
+AVAILABLE_TOOLS = [
+    "CodeDocsSearchTool", "CSVSearchTool", "DirectorySearchTool", "DOCXSearchTool",
+    "DirectoryReadTool", "FileReadTool", "TXTSearchTool", "JSONSearchTool",
+    "MDXSearchTool", "PDFSearchTool", "RagTool", "ScrapeElementFromWebsiteTool",
+    "ScrapeWebsiteTool", "WebsiteSearchTool", "XMLSearchTool",
+    "YoutubeChannelSearchTool", "YoutubeVideoSearchTool"
+]
+
+# =============================================================================
+# Base Generator Class (DRY - shared functionality)
+# =============================================================================
+class BaseAutoGenerator:
+    """
+    Base class for auto-generators with shared functionality.
+    
+    Provides:
+    - Lazy-loaded instructor client (LiteLLM or OpenAI fallback)
+    - Environment variable handling for model/API configuration
+    - Config list management
+    """
+    
+    def __init__(self, config_list: Optional[List[Dict]] = None):
+        """
+        Initialize base generator with LLM configuration.
+        
+        Args:
+            config_list: Optional LLM configuration list
+        """
+        # Support multiple environment variable patterns for better compatibility
+        model_name = os.environ.get("MODEL_NAME") or os.environ.get("OPENAI_MODEL_NAME", "gpt-4o-mini")
+        base_url = (
+            os.environ.get("OPENAI_BASE_URL") or 
+            os.environ.get("OPENAI_API_BASE") or
+            os.environ.get("OLLAMA_API_BASE", "https://api.openai.com/v1")
+        )
+        
+        self.config_list = config_list or [
+            {
+                'model': model_name,
+                'base_url': base_url,
+                'api_key': os.environ.get("OPENAI_API_KEY")
+            }
+        ]
+        self._client = None  # Lazy loading for performance
+    
+    @property
+    def client(self):
+        """Lazy load the instructor client to avoid performance impact.
+        
+        Uses LiteLLM via instructor.from_provider if available for multi-provider support,
+        otherwise falls back to direct OpenAI SDK.
+        """
+        if self._client is None:
+            if LITELLM_AVAILABLE:
+                # Use LiteLLM for multi-provider support (100+ LLMs)
+                model_name = self.config_list[0]['model']
+                self._client = instructor.from_provider(
+                    f"litellm/{model_name}",
+                    mode=instructor.Mode.JSON
+                )
+            else:
+                # Fallback to direct OpenAI SDK
+                self._client = instructor.patch(
+                    OpenAI(
+                        base_url=self.config_list[0]['base_url'],
+                        api_key=self.config_list[0]['api_key'],
+                    ),
+                    mode=instructor.Mode.JSON,
+                )
+        return self._client
+    
+    @staticmethod
+    def get_available_tools() -> List[str]:
+        """Return list of available tools for agent assignment."""
+        return AVAILABLE_TOOLS.copy()
+    
+    @staticmethod
+    def analyze_complexity(topic: str) -> str:
+        """
+        Analyze task complexity based on keywords.
+        
+        Args:
+            topic: The task description
+            
+        Returns:
+            str: Complexity level - 'simple', 'moderate', or 'complex'
+        """
+        topic_lower = topic.lower()
+        
+        # Complex task indicators
+        complex_keywords = [
+            'comprehensive', 'multi-step', 'analyze and', 'research and write',
+            'multiple', 'coordinate', 'complex', 'detailed analysis',
+            'full report', 'in-depth', 'thorough'
+        ]
+        
+        # Simple task indicators
+        simple_keywords = [
+            'write a', 'create a', 'simple', 'quick', 'brief',
+            'haiku', 'poem', 'summary', 'list', 'single'
+        ]
+        
+        if any(kw in topic_lower for kw in complex_keywords):
+            return 'complex'
+        elif any(kw in topic_lower for kw in simple_keywords):
+            return 'simple'
+        else:
+            return 'moderate'
+
+
+# =============================================================================
+# Pydantic Models for Structured Output
+# =============================================================================
+
 class TaskDetails(BaseModel):
+    """Details for a single task."""
    description: str
    expected_output: str

 class RoleDetails(BaseModel):
+    """Details for a single role/agent."""
    role: str
    goal: str
    backstory: str
@@ -66,14 +192,62 @@ class RoleDetails(BaseModel):
    tools: List[str]

 class TeamStructure(BaseModel):
+    """Structure for multi-agent team."""
    roles: Dict[str, RoleDetails]

-class AutoGenerator:
-    def __init__(self, topic="Movie Story writing about AI", agent_file="test.yaml", framework="crewai", config_list: Optional[List[Dict]] = None):
+class SingleAgentStructure(BaseModel):
+    """Structure for single-agent generation (Anthropic's 'start simple' principle)."""
+    name: str
+    role: str
+    goal: str
+    backstory: str
+    instructions: str
+    tools: List[str] = []
+    task_description: str
+    expected_output: str
+
+class PatternRecommendation(BaseModel):
+    """LLM-based pattern recommendation with reasoning."""
+    pattern: str  # sequential, parallel, routing, orchestrator-workers, evaluator-optimizer
+    reasoning: str  # Why this pattern was chosen
+    confidence: float  # 0.0 to 1.0 confidence score
+
+class ValidationGate(BaseModel):
+    """Validation gate for prompt chaining workflows."""
+    criteria: str  # What to validate
+    pass_action: str  # Action if validation passes (e.g., "continue", "next_step")
+    fail_action: str  # Action if validation fails (e.g., "retry", "escalate", "abort")
+
+class AutoGenerator(BaseAutoGenerator):
+    """
+    Auto-generates agents.yaml files from a topic description.
+    
+    Inherits from BaseAutoGenerator for shared LLM client functionality.
+    
+    Usage:
+        generator = AutoGenerator(framework="crewai", topic="Create a movie script")
+        path = generator.generate()
+    """
+    
+    def __init__(self, topic="Movie Story writing about AI", agent_file="test.yaml", 
+                 framework="crewai", config_list: Optional[List[Dict]] = None,
+                 pattern: str = "sequential", single_agent: bool = False):
        """
        Initialize the AutoGenerator class with the specified topic, agent file, and framework.
+        
+        Args:
+            topic: The task/topic for agent generation
+            agent_file: Output YAML file name
+            framework: Framework to use (crewai, autogen, praisonai)
+            config_list: Optional LLM configuration
+            pattern: Workflow pattern (sequential, parallel, routing, orchestrator-workers, evaluator-optimizer)
+            single_agent: If True, generate a single agent instead of a team
+        
        Note: autogen framework is different from this AutoGenerator class.
        """
+        # Initialize base class first (handles config_list and client)
+        super().__init__(config_list=config_list)
+        
        # Validate framework availability and show framework-specific messages
        if framework == "crewai" and not CREWAI_AVAILABLE:
            raise ImportError("""
@@ -106,35 +280,42 @@ Tools are not available for {framework}. To use tools, install:
    pip install "praisonai[{framework}]"
 """)

-        # Support multiple environment variable patterns for better compatibility
-        # Priority order: MODEL_NAME > OPENAI_MODEL_NAME for model selection
-        model_name = os.environ.get("MODEL_NAME") or os.environ.get("OPENAI_MODEL_NAME", "gpt-5-nano")
-        
-        # Priority order for base_url: OPENAI_BASE_URL > OPENAI_API_BASE > OLLAMA_API_BASE
-        # OPENAI_BASE_URL is the standard OpenAI SDK environment variable
-        base_url = (
-            os.environ.get("OPENAI_BASE_URL") or 
-            os.environ.get("OPENAI_API_BASE") or
-            os.environ.get("OLLAMA_API_BASE", "https://api.openai.com/v1")
-        )
-        
-        self.config_list = config_list or [
-            {
-                'model': model_name,
-                'base_url': base_url,
-                'api_key': os.environ.get("OPENAI_API_KEY")
-            }
-        ]
        self.topic = topic
        self.agent_file = agent_file
        self.framework = framework or "praisonai"
-        self.client = instructor.patch(
-            OpenAI(
-                base_url=self.config_list[0]['base_url'],
-                api_key=self.config_list[0]['api_key'],
-            ),
-            mode=instructor.Mode.JSON,
-        )
+        self.pattern = pattern
+        self.single_agent = single_agent
+    
+    def recommend_pattern(self, topic: str = None) -> str:
+        """
+        Recommend the best workflow pattern based on task characteristics.
+        
+        Args:
+            topic: The task description (uses self.topic if not provided)
+            
+        Returns:
+            str: Recommended pattern name
+        """
+        task = topic or self.topic
+        task_lower = task.lower()
+        
+        # Keywords that suggest specific patterns
+        parallel_keywords = ['multiple', 'concurrent', 'parallel', 'simultaneously', 'different sources', 'compare', 'various']
+        routing_keywords = ['classify', 'categorize', 'route', 'different types', 'depending on', 'if...then']
+        orchestrator_keywords = ['complex', 'comprehensive', 'multi-step', 'coordinate', 'delegate', 'break down', 'analyze and']
+        evaluator_keywords = ['refine', 'improve', 'iterate', 'quality', 'review', 'feedback', 'polish', 'optimize']
+        
+        # Check for pattern indicators
+        if any(kw in task_lower for kw in evaluator_keywords):
+            return "evaluator-optimizer"
+        elif any(kw in task_lower for kw in orchestrator_keywords):
+            return "orchestrator-workers"
+        elif any(kw in task_lower for kw in routing_keywords):
+            return "routing"
+        elif any(kw in task_lower for kw in parallel_keywords):
+            return "parallel"
+        else:
+            return "sequential"

    def generate(self, merge=False):
        """
@@ -157,7 +338,8 @@ Tools are not available for {framework}. To use tools, install:
        response = self.client.chat.completions.create(
            model=self.config_list[0]['model'],
            response_model=TeamStructure,
-            max_retries=10,
+            max_retries=5,
+            timeout=120.0,  # 2 minute timeout for complex generations
            messages=[
                {"role": "system", "content": "You are a helpful assistant designed to output complex team structures."},
                {"role": "user", "content": self.get_user_content()}
@@ -194,8 +376,7 @@ Tools are not available for {framework}. To use tools, install:
                    "goal": role_details['goal'],
                    "role": role_details['role'],
                    "tasks": {},
-                    # "tools": role_details.get('tools', []),
-                    "tools": ['']
+                    "tools": role_details.get('tools', [])
                }

                for task_id, task_details in role_details['tasks'].items():
@@ -265,7 +446,7 @@ Tools are not available for {framework}. To use tools, install:
                "goal": role_details['goal'],
                "role": role_details['role'],
                "tasks": {},
-                "tools": ['']
+                "tools": role_details.get('tools', [])
            }
            
            # Add tasks for this role
@@ -292,45 +473,77 @@ Tools are not available for {framework}. To use tools, install:
            prompt = generator.get_user_content()
            print(prompt)
        """
-        user_content = """Generate a team structure for  \"""" + self.topic + """\" task. 
-No Input data will be provided to the team.
-The team will work in sequence. First role will pass the output to the next role, and so on.
-The last role will generate the final output.
-Think step by step.
-With maximum 3 roles, each with 1 task. Include role goals, backstories, task descriptions, and expected outputs.
-List of Available Tools: CodeDocsSearchTool, CSVSearchTool, DirectorySearchTool, DOCXSearchTool, DirectoryReadTool, FileReadTool, TXTSearchTool, JSONSearchTool, MDXSearchTool, PDFSearchTool, RagTool, ScrapeElementFromWebsiteTool, ScrapeWebsiteTool, WebsiteSearchTool, XMLSearchTool, YoutubeChannelSearchTool, YoutubeVideoSearchTool.
-Only use Available Tools. Do Not use any other tools. 
-Example Below: 
-Use below example to understand the structure of the output. 
-The final role you create should satisfy the provided task: """ + self.topic + """.
-{
-"roles": {
-"narrative_designer": {
-"role": "Narrative Designer",
-"goal": "Create AI storylines",
-"backstory": "Skilled in narrative development for AI, with a focus on story resonance.",
-"tools": ["ScrapeWebsiteTool"],
-"tasks": {
-"story_concept_development": {
-"description": "Craft a unique AI story concept with depth and engagement using concept from this page the content https://www.asthebirdfliesblog.com/posts/how-to-write-book-story-development .",
-"expected_output": "Document with narrative arcs, character bios, and settings."
-}
-}
-},
-"scriptwriter": {
-"role": "Scriptwriter",
-"goal": "Write scripts from AI concepts",
-"backstory": "Expert in dialogue and script structure, translating concepts into scripts.",
-"tasks": {
-"scriptwriting_task": {
-"description": "Turn narrative concepts into scripts, including dialogue and scenes.",
-"expected_output": "Production-ready script with dialogue and scene details."
-}
-}
-}
-}
-}
-        """
+        # Pattern-specific guidance
+        pattern_guidance = {
+            "sequential": "The team will work in sequence. Each role passes output to the next.",
+            "parallel": "The team will work in parallel on independent subtasks, then combine results.",
+            "routing": "A classifier agent will route requests to specialized agents based on input type.",
+            "orchestrator-workers": "A central orchestrator will dynamically delegate tasks to specialized workers.",
+            "evaluator-optimizer": "One agent generates content, another evaluates it in a loop until quality criteria are met."
+        }
+        
+        workflow_guidance = pattern_guidance.get(self.pattern, pattern_guidance["sequential"])
+        
+        user_content = f"""Analyze and generate a team structure for: "{self.topic}"
+
+STEP 1: TASK COMPLEXITY ANALYSIS
+First, analyze the task complexity:
+- Is this a simple task that 1-2 agents could handle?
+- Does it require multiple specialized perspectives?
+- Are there clear sequential dependencies between steps?
+
+STEP 2: DETERMINE OPTIMAL TEAM SIZE
+Based on your analysis, create the minimum number of agents needed:
+- Simple tasks: 1-2 agents
+- Moderate tasks: 2-3 agents  
+- Complex tasks: 3-4 agents (maximum)
+
+IMPORTANT: Only create agents that provide meaningful specialization. Avoid unnecessary complexity.
+
+STEP 3: DESIGN THE TEAM (Pattern: {self.pattern})
+{workflow_guidance}
+Each agent should have:
+- A clear, distinct role
+- A specific goal
+- Relevant backstory
+- 1 focused task with clear description and expected output
+- Appropriate tools (only if needed)
+
+Available Tools: CodeDocsSearchTool, CSVSearchTool, DirectorySearchTool, DOCXSearchTool, DirectoryReadTool, FileReadTool, TXTSearchTool, JSONSearchTool, MDXSearchTool, PDFSearchTool, RagTool, ScrapeElementFromWebsiteTool, ScrapeWebsiteTool, WebsiteSearchTool, XMLSearchTool, YoutubeChannelSearchTool, YoutubeVideoSearchTool.
+Only use these tools if the task requires them. Use empty list [] if no tools needed.
+
+Example structure (2 agents for a writing task):
+{{
+  "roles": {{
+    "researcher": {{
+      "role": "Research Analyst",
+      "goal": "Gather comprehensive information on the topic",
+      "backstory": "Expert researcher skilled at finding and synthesizing information.",
+      "tools": ["WebsiteSearchTool"],
+      "tasks": {{
+        "research_task": {{
+          "description": "Research key information about the topic and compile findings.",
+          "expected_output": "Comprehensive research notes with key facts and insights."
+        }}
+      }}
+    }},
+    "writer": {{
+      "role": "Content Writer",
+      "goal": "Create polished final content",
+      "backstory": "Skilled writer who transforms research into engaging content.",
+      "tools": [],
+      "tasks": {{
+        "writing_task": {{
+          "description": "Write the final content based on research findings.",
+          "expected_output": "Polished, well-structured final document."
+        }}
+      }}
+    }}
+  }}
+}}
+
+Now generate the optimal team structure for: {self.topic}
+"""
        return user_content

    
@@ -372,13 +585,14 @@ class WorkflowStructure(BaseModel):
    description: str
    agents: Dict[str, WorkflowAgentDetails]
    steps: List[Dict]  # Can be agent steps, route, parallel, etc.
+    gates: Optional[List[ValidationGate]] = None  # Optional validation gates


-class WorkflowAutoGenerator:
+class WorkflowAutoGenerator(BaseAutoGenerator):
    """
    Auto-generates workflow.yaml files from a topic description.
    
-    Uses lazy loading to avoid performance impact when not used.
+    Inherits from BaseAutoGenerator for shared LLM client functionality.
    
    Usage:
        generator = WorkflowAutoGenerator(topic="Research AI trends and write a report")
@@ -387,7 +601,9 @@ class WorkflowAutoGenerator:
    
    def __init__(self, topic: str = "Research and write about AI", 
                 workflow_file: str = "workflow.yaml",
-                 config_list: Optional[List[Dict]] = None):
+                 config_list: Optional[List[Dict]] = None,
+                 framework: str = "praisonai",
+                 single_agent: bool = False):
        """
        Initialize the WorkflowAutoGenerator.
        
@@ -395,45 +611,105 @@ class WorkflowAutoGenerator:
            topic: The task/topic for the workflow
            workflow_file: Output file name
            config_list: Optional LLM configuration
+            framework: Framework to use (praisonai, crewai, autogen)
+            single_agent: If True, generate a single agent workflow
        """
-        # Support multiple environment variable patterns
-        model_name = os.environ.get("MODEL_NAME") or os.environ.get("OPENAI_MODEL_NAME", "gpt-4o-mini")
-        base_url = (
-            os.environ.get("OPENAI_BASE_URL") or 
-            os.environ.get("OPENAI_API_BASE") or
-            os.environ.get("OLLAMA_API_BASE", "https://api.openai.com/v1")
-        )
+        # Initialize base class (handles config_list and client)
+        super().__init__(config_list=config_list)
        
-        self.config_list = config_list or [
-            {
-                'model': model_name,
-                'base_url': base_url,
-                'api_key': os.environ.get("OPENAI_API_KEY")
-            }
-        ]
        self.topic = topic
        self.workflow_file = workflow_file
-        self._client = None  # Lazy loading
+        self.framework = framework
+        self.single_agent = single_agent
    
-    @property
-    def client(self):
-        """Lazy load the OpenAI client to avoid performance impact."""
-        if self._client is None:
-            self._client = instructor.patch(
-                OpenAI(
-                    base_url=self.config_list[0]['base_url'],
-                    api_key=self.config_list[0]['api_key'],
-                ),
-                mode=instructor.Mode.JSON,
-            )
-        return self._client
+    def recommend_pattern(self, topic: str = None) -> str:
+        """
+        Recommend the best workflow pattern based on task characteristics.
+        
+        Args:
+            topic: The task description (uses self.topic if not provided)
+            
+        Returns:
+            str: Recommended pattern name
+            
+        Pattern recommendations based on Anthropic's best practices:
+        - sequential: Clear step-by-step dependencies
+        - parallel: Independent subtasks that can run concurrently
+        - routing: Different input types need different handling
+        - orchestrator-workers: Complex tasks needing dynamic decomposition
+        - evaluator-optimizer: Tasks requiring iterative refinement
+        """
+        task = topic or self.topic
+        task_lower = task.lower()
+        
+        # Keywords that suggest specific patterns
+        parallel_keywords = ['multiple', 'concurrent', 'parallel', 'simultaneously', 'different sources', 'compare', 'various']
+        routing_keywords = ['classify', 'categorize', 'route', 'different types', 'depending on', 'if...then']
+        orchestrator_keywords = ['complex', 'comprehensive', 'multi-step', 'coordinate', 'delegate', 'break down', 'analyze and']
+        evaluator_keywords = ['refine', 'improve', 'iterate', 'quality', 'review', 'feedback', 'polish', 'optimize']
+        
+        # Check for pattern indicators
+        if any(kw in task_lower for kw in evaluator_keywords):
+            return "evaluator-optimizer"
+        elif any(kw in task_lower for kw in orchestrator_keywords):
+            return "orchestrator-workers"
+        elif any(kw in task_lower for kw in routing_keywords):
+            return "routing"
+        elif any(kw in task_lower for kw in parallel_keywords):
+            return "parallel"
+        else:
+            return "sequential"
    
-    def generate(self, pattern: str = "sequential") -> str:
+    def recommend_pattern_llm(self, topic: str = None) -> PatternRecommendation:
+        """
+        Use LLM to recommend the best workflow pattern with reasoning.
+        
+        Args:
+            topic: The task description (uses self.topic if not provided)
+            
+        Returns:
+            PatternRecommendation: Pattern with reasoning and confidence score
+        """
+        task = topic or self.topic
+        
+        prompt = f"""Analyze this task and recommend the best workflow pattern:
+
+Task: "{task}"
+
+Available patterns:
+1. sequential - Agents work one after another, passing output to the next
+2. parallel - Multiple agents work concurrently on independent subtasks
+3. routing - A classifier routes requests to specialized agents based on input type
+4. orchestrator-workers - Central orchestrator dynamically delegates to specialized workers
+5. evaluator-optimizer - Generator creates content, evaluator reviews in a loop until quality met
+
+Respond with:
+- pattern: The recommended pattern name
+- reasoning: Why this pattern is best for this task
+- confidence: Your confidence score (0.0 to 1.0)
+"""
+        
+        response = self.client.chat.completions.create(
+            model=self.config_list[0]['model'],
+            response_model=PatternRecommendation,
+            max_retries=3,
+            timeout=60.0,
+            messages=[
+                {"role": "system", "content": "You are an expert at designing AI agent workflows."},
+                {"role": "user", "content": prompt}
+            ]
+        )
+        
+        return response
+    
+    def generate(self, pattern: str = "sequential", merge: bool = False) -> str:
        """
        Generate a workflow YAML file.
        
        Args:
-            pattern: Workflow pattern - "sequential", "routing", "parallel", "loop"
+            pattern: Workflow pattern - "sequential", "routing", "parallel", "loop",
+                     "orchestrator-workers", "evaluator-optimizer"
+            merge: If True, merge with existing workflow file instead of overwriting
            
        Returns:
            Path to the generated workflow file
@@ -442,6 +718,7 @@ class WorkflowAutoGenerator:
            model=self.config_list[0]['model'],
            response_model=WorkflowStructure,
            max_retries=5,
+            timeout=120.0,  # 2 minute timeout for complex generations
            messages=[
                {"role": "system", "content": "You are a helpful assistant that designs workflow structures."},
                {"role": "user", "content": self._get_prompt(pattern)}
@@ -449,16 +726,86 @@ class WorkflowAutoGenerator:
        )
        
        json_data = json.loads(response.model_dump_json())
+        
+        if merge and os.path.exists(self.workflow_file):
+            return self._save_workflow(self.merge_with_existing_workflow(json_data), pattern)
        return self._save_workflow(json_data, pattern)
    
+    def merge_with_existing_workflow(self, new_data: Dict) -> Dict:
+        """
+        Merge new workflow data with existing workflow file.
+        
+        Args:
+            new_data: The new workflow data to merge
+            
+        Returns:
+            Dict: Merged workflow data
+        """
+        try:
+            with open(self.workflow_file, 'r') as f:
+                existing_data = yaml.safe_load(f)
+            
+            if not existing_data:
+                return new_data
+        except (yaml.YAMLError, FileNotFoundError) as e:
+            logging.warning(f"Could not load existing workflow file {self.workflow_file}: {e}")
+            return new_data
+        
+        # Merge agents (avoid duplicates)
+        merged_agents = existing_data.get('agents', {}).copy()
+        for agent_id, agent_data in new_data.get('agents', {}).items():
+            # Rename if conflict
+            final_id = agent_id
+            counter = 1
+            while final_id in merged_agents:
+                final_id = f"{agent_id}_auto_{counter}"
+                counter += 1
+            merged_agents[final_id] = agent_data
+        
+        # Merge steps (append new steps)
+        merged_steps = existing_data.get('steps', []) + new_data.get('steps', [])
+        
+        # Create merged structure
+        merged = {
+            'name': existing_data.get('name', new_data.get('name', 'Merged Workflow')),
+            'description': f"{existing_data.get('description', '')} + {new_data.get('description', '')}",
+            'agents': merged_agents,
+            'steps': merged_steps
+        }
+        
+        return merged
+    
    def _get_prompt(self, pattern: str) -> str:
        """Generate the prompt based on the workflow pattern."""
+        # Analyze complexity to determine agent count
+        complexity = self.analyze_complexity(self.topic)
+        if complexity == 'simple':
+            agent_guidance = "Create 1-2 agents (simple task detected)."
+        elif complexity == 'complex':
+            agent_guidance = "Create 3-4 agents (complex task detected)."
+        else:
+            agent_guidance = "Create 2-3 agents (moderate task detected)."
+        
+        # Get available tools
+        tools_list = ", ".join(self.get_available_tools())
+        
        base_prompt = f"""Generate a workflow structure for: "{self.topic}"

+STEP 1: ANALYZE TASK COMPLEXITY
+- Is this a simple task (1-2 agents)?
+- Does it require multiple specialists (2-3 agents)?
+- Is it complex with many dependencies (3-4 agents)?
+
+STEP 2: DESIGN WORKFLOW
 The workflow should use the "{pattern}" pattern.
-Create 2-4 agents with clear roles and instructions.
+{agent_guidance}
+Each agent should have clear roles and instructions.
 Each step should have a clear action.

+STEP 3: ASSIGN TOOLS (if needed)
+Available Tools: {tools_list}
+Only assign tools if the task requires them. Use empty list or null if no tools needed.
+
 """
        
        if pattern == "routing":
@@ -501,6 +848,59 @@ Example structure:
    {"agent": "aggregator", "action": "Combine all findings"}
  ]
 }
+"""
+        elif pattern == "orchestrator-workers":
+            base_prompt += """
+Create an orchestrator-workers workflow where a central orchestrator dynamically delegates tasks to specialized workers.
+The orchestrator analyzes the input, decides which workers are needed, and synthesizes results.
+
+Example structure:
+{
+  "name": "Orchestrator-Workers Workflow",
+  "description": "Central orchestrator delegates to specialized workers",
+  "agents": {
+    "orchestrator": {"name": "Orchestrator", "role": "Task Coordinator", "goal": "Analyze tasks and delegate to appropriate workers", "instructions": "Break down the task, identify required specialists, and coordinate their work. Output a JSON with 'subtasks' array listing which workers to invoke."},
+    "researcher": {"name": "Researcher", "role": "Research Specialist", "goal": "Gather information", "instructions": "Research and provide factual information"},
+    "analyst": {"name": "Analyst", "role": "Data Analyst", "goal": "Analyze data and patterns", "instructions": "Analyze information and identify insights"},
+    "writer": {"name": "Writer", "role": "Content Writer", "goal": "Create written content", "instructions": "Write clear, engaging content"},
+    "synthesizer": {"name": "Synthesizer", "role": "Results Synthesizer", "goal": "Combine all worker outputs", "instructions": "Synthesize all worker outputs into a coherent final result"}
+  },
+  "steps": [
+    {"agent": "orchestrator", "action": "Analyze task and determine required workers: {{input}}"},
+    {"name": "worker_dispatch", "parallel": [
+      {"agent": "researcher", "action": "Research: {{input}}"},
+      {"agent": "analyst", "action": "Analyze: {{input}}"},
+      {"agent": "writer", "action": "Draft content for: {{input}}"}
+    ]},
+    {"agent": "synthesizer", "action": "Combine all worker outputs into final result"}
+  ]
+}
+"""
+        elif pattern == "evaluator-optimizer":
+            base_prompt += """
+Create an evaluator-optimizer workflow where one agent generates content and another evaluates it in a loop.
+The generator improves based on evaluator feedback until quality criteria are met.
+
+Example structure:
+{
+  "name": "Evaluator-Optimizer Workflow",
+  "description": "Iterative refinement through generation and evaluation",
+  "agents": {
+    "generator": {"name": "Generator", "role": "Content Generator", "goal": "Generate high-quality content", "instructions": "Create content based on the input. If feedback is provided, improve the content accordingly."},
+    "evaluator": {"name": "Evaluator", "role": "Quality Evaluator", "goal": "Evaluate content quality", "instructions": "Evaluate the content on: clarity, accuracy, completeness, and relevance. Score 1-10 for each. If average score < 7, provide specific improvement feedback. If score >= 7, respond with 'APPROVED'."}
+  },
+  "steps": [
+    {"agent": "generator", "action": "Generate initial content for: {{input}}"},
+    {"name": "evaluation_loop", "loop": {
+      "agent": "evaluator",
+      "action": "Evaluate the generated content",
+      "condition": "output does not contain 'APPROVED'",
+      "max_iterations": 3,
+      "feedback_to": "generator"
+    }},
+    {"agent": "generator", "action": "Finalize content based on all feedback"}
+  ]
+}
 """
        else:  # sequential
            base_prompt += """
@@ -8,12 +8,19 @@ Structure:
 - features/: Feature handlers for CLI flags and commands
 """

-from .main import PraisonAI
-
 __all__ = ["PraisonAI"]


+def __getattr__(name):
+    """Lazy load PraisonAI to avoid slow imports."""
+    if name == "PraisonAI":
+        from .main import PraisonAI
+        return PraisonAI
+    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
+
+
 def main():
    """CLI entry point function."""
+    from .main import PraisonAI
    praison_ai = PraisonAI()
    praison_ai.main()
@@ -21,6 +21,14 @@ Features:
 - flow_display: Visual workflow tracking
 - workflow: YAML workflow management
 - n8n: Export workflows to n8n visual editor
+- external_agents: External AI CLI tool integrations (Claude, Gemini, Codex, Cursor)
+- slash_commands: Interactive slash commands (/help, /cost, /plan, etc.)
+- autonomy_mode: Autonomy levels (suggest, auto-edit, full-auto)
+- cost_tracker: Real-time cost and token tracking
+- message_queue: Message queuing while agent is processing
+- at_mentions: @ mention autocomplete for files and directories
+- compare: CLI command comparison feature
+- eval: Agent evaluation framework (accuracy, performance, reliability, criteria)
 """

 # Type hints available for IDE support
@@ -43,6 +51,26 @@ __all__ = [
    'FlowDisplayHandler',
    'WorkflowHandler',
    'N8nHandler',
+    'ExternalAgentsHandler',
+    # New CLI enhancement features
+    'SlashCommandHandler',
+    'AutonomyModeHandler',
+    'CostTrackerHandler',
+    'RepoMapHandler',
+    'InteractiveTUIHandler',
+    'GitIntegrationHandler',
+    'SandboxExecutorHandler',
+    'MessageQueueHandler',
+    # @ mention autocomplete
+    'AtMentionCompleter',
+    'CombinedCompleter',
+    'FileSearchService',
+    # Evaluation
+    'EvalHandler',
+    # Compare feature
+    'CompareHandler',
+    # Agent Skills
+    'SkillsHandler',
 ]

 def __getattr__(name):
@@ -95,4 +123,51 @@ def __getattr__(name):
    elif name == 'N8nHandler':
        from .n8n import N8nHandler
        return N8nHandler
+    elif name == 'ExternalAgentsHandler':
+        from .external_agents import ExternalAgentsHandler
+        return ExternalAgentsHandler
+    # New CLI enhancement features
+    elif name == 'SlashCommandHandler':
+        from .slash_commands import SlashCommandHandler
+        return SlashCommandHandler
+    elif name == 'AutonomyModeHandler':
+        from .autonomy_mode import AutonomyModeHandler
+        return AutonomyModeHandler
+    elif name == 'CostTrackerHandler':
+        from .cost_tracker import CostTrackerHandler
+        return CostTrackerHandler
+    elif name == 'RepoMapHandler':
+        from .repo_map import RepoMapHandler
+        return RepoMapHandler
+    elif name == 'InteractiveTUIHandler':
+        from .interactive_tui import InteractiveTUIHandler
+        return InteractiveTUIHandler
+    elif name == 'GitIntegrationHandler':
+        from .git_integration import GitIntegrationHandler
+        return GitIntegrationHandler
+    elif name == 'SandboxExecutorHandler':
+        from .sandbox_executor import SandboxExecutorHandler
+        return SandboxExecutorHandler
+    elif name == 'MessageQueueHandler':
+        from .message_queue import MessageQueueHandler
+        return MessageQueueHandler
+    # @ mention autocomplete
+    elif name == 'AtMentionCompleter':
+        from .at_mentions import AtMentionCompleter
+        return AtMentionCompleter
+    elif name == 'CombinedCompleter':
+        from .at_mentions import CombinedCompleter
+        return CombinedCompleter
+    elif name == 'FileSearchService':
+        from .at_mentions import FileSearchService
+        return FileSearchService
+    elif name == 'CompareHandler':
+        from .compare import CompareHandler
+        return CompareHandler
+    elif name == 'EvalHandler':
+        from .eval import EvalHandler
+        return EvalHandler
+    elif name == 'SkillsHandler':
+        from .skills import SkillsHandler
+        return SkillsHandler
    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
@@ -0,0 +1,734 @@
+"""
+Agent Scheduler CLI Feature.
+
+Provides CLI commands for scheduling agents to run 24/7 at regular intervals.
+"""
+
+import logging
+
+logger = logging.getLogger(__name__)
+
+
+class AgentSchedulerHandler:
+    """Handler for agent scheduler CLI commands."""
+    
+    @staticmethod
+    def handle_daemon_command(subcommand: str, args, unknown_args=None) -> int:
+        """
+        Handle daemon management commands (start, list, stop, logs, restart).
+        
+        Args:
+            subcommand: Command to execute (start, list, stop, logs, restart)
+            args: Parsed command-line arguments
+            unknown_args: Additional arguments
+            
+        Returns:
+            Exit code
+        """
+        from praisonai.scheduler.state_manager import SchedulerStateManager
+        from praisonai.scheduler.daemon_manager import DaemonManager
+        
+        state_manager = SchedulerStateManager()
+        daemon_manager = DaemonManager()
+        
+        if subcommand == "start":
+            return AgentSchedulerHandler._handle_start(args, unknown_args, state_manager, daemon_manager)
+        elif subcommand == "list":
+            return AgentSchedulerHandler._handle_list(state_manager)
+        elif subcommand == "stop":
+            return AgentSchedulerHandler._handle_stop(unknown_args, state_manager, daemon_manager)
+        elif subcommand == "logs":
+            return AgentSchedulerHandler._handle_logs(unknown_args, daemon_manager)
+        elif subcommand == "restart":
+            return AgentSchedulerHandler._handle_restart(unknown_args, state_manager, daemon_manager)
+        elif subcommand == "delete":
+            return AgentSchedulerHandler._handle_delete(unknown_args, state_manager)
+        elif subcommand == "describe":
+            return AgentSchedulerHandler._handle_describe(unknown_args, state_manager, daemon_manager)
+        elif subcommand == "save":
+            return AgentSchedulerHandler._handle_save(unknown_args, state_manager)
+        elif subcommand == "stop-all":
+            return AgentSchedulerHandler._handle_stop_all(state_manager, daemon_manager)
+        elif subcommand == "stats":
+            return AgentSchedulerHandler._handle_stats(state_manager, unknown_args, daemon_manager)
+        else:
+            print(f"❌ Unknown subcommand: {subcommand}")
+            print("\nAvailable commands:")
+            print("  start, list, stop, logs, restart, delete, describe, save, stop-all, stats")
+            return 1
+    
+    @staticmethod
+    def _handle_start(args, unknown_args, state_manager, daemon_manager) -> int:
+        """Handle 'schedule start' command."""
+        from datetime import datetime
+        
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name and task")
+            print("\nUsage:")
+            print('  praisonai schedule start <name> "Your task" --interval hourly')
+            print("\nExample:")
+            print('  praisonai schedule start news-checker "Check AI news" --interval hourly')
+            return 1
+        
+        # Parse arguments
+        name = unknown_args[0]
+        task = unknown_args[1] if len(unknown_args) > 1 else None
+        
+        if not task:
+            print("❌ Error: Please provide a task")
+            return 1
+        
+        # Get options
+        interval = getattr(args, 'schedule_interval', None) or 'hourly'
+        max_retries = getattr(args, 'schedule_max_retries', None) or 3
+        timeout = getattr(args, 'timeout', None)
+        max_cost = getattr(args, 'max_cost', None)
+        
+        # Check if name already exists
+        existing = state_manager.load_state(name)
+        if existing and daemon_manager.get_status(existing.get('pid', 0))['is_alive']:
+            print(f"❌ Error: Scheduler '{name}' is already running (PID: {existing['pid']})")
+            print(f"   Use 'praisonai schedule stop {name}' to stop it first")
+            return 1
+        
+        # Start daemon
+        print(f"🚀 Starting scheduler '{name}'...")
+        print(f"   Task: {task}")
+        print(f"   Interval: {interval}")
+        if timeout:
+            print(f"   Timeout: {timeout}s")
+        if max_cost:
+            print(f"   Budget: ${max_cost}")
+        
+        pid = daemon_manager.start_scheduler_daemon(
+            name=name,
+            task=task,
+            interval=interval,
+            max_cost=max_cost,
+            timeout=timeout,
+            max_retries=max_retries
+        )
+        
+        # Save state
+        state = {
+            "name": name,
+            "pid": pid,
+            "task": task,
+            "interval": interval,
+            "timeout": timeout,
+            "max_cost": max_cost,
+            "max_retries": max_retries,
+            "status": "running",
+            "started_at": datetime.now().isoformat(),
+            "executions": 0,
+            "cost": 0.0
+        }
+        state_manager.save_state(name, state)
+        
+        print(f"\n✅ Scheduler '{name}' started successfully!")
+        print(f"   PID: {pid}")
+        print(f"   Logs: ~/.praisonai/logs/{name}.log")
+        print(f"\nManage:")
+        print(f"   praisonai schedule list")
+        print(f"   praisonai schedule describe {name}")
+        print(f"   praisonai schedule logs {name} --follow")
+        print(f"   praisonai schedule stop {name}")
+        
+        return 0
+    
+    @staticmethod
+    def _handle_list(state_manager) -> int:
+        """Handle 'schedule list' command."""
+        states = state_manager.list_all()
+        
+        if not states:
+            print("No schedulers running.")
+            print("\nStart one with:")
+            print('  praisonai schedule start <name> "Your task" --interval hourly')
+            return 0
+        
+        # Clean up dead processes
+        state_manager.cleanup_dead_processes()
+        states = state_manager.list_all()
+        
+        # Display table
+        print(f"\n{'Name':<20} {'Status':<10} {'PID':<8} {'Interval':<12} {'Task':<40}")
+        print("=" * 90)
+        
+        for state in states:
+            name = state.get('name', 'unknown')[:20]
+            pid = state.get('pid', 0)
+            status = "running" if state_manager.is_process_alive(pid) else "stopped"
+            interval = state.get('interval', 'unknown')[:12]
+            task = state.get('task', '')[:40]
+            
+            status_icon = "🟢" if status == "running" else "🔴"
+            print(f"{name:<20} {status_icon} {status:<8} {pid:<8} {interval:<12} {task:<40}")
+        
+        print(f"\nTotal: {len(states)} scheduler(s)")
+        return 0
+    
+    @staticmethod
+    def _handle_stats(state_manager, unknown_args=None, daemon_manager=None) -> int:
+        """Handle 'schedule stats' command - show aggregate or individual statistics."""
+        # If a name is provided, show individual stats (alias for describe)
+        if unknown_args and len(unknown_args) > 0:
+            name = unknown_args[0]
+            return AgentSchedulerHandler._handle_describe([name], state_manager, daemon_manager)
+        
+        # Otherwise show aggregate stats
+        states = state_manager.list_all()
+        
+        if not states:
+            print("No schedulers found.")
+            print("\nStart one with:")
+            print('  praisonai schedule start <name> "Your task" --interval hourly')
+            return 0
+        
+        # Calculate aggregate stats
+        total_schedulers = len(states)
+        running = sum(1 for s in states if s.get('status') == 'running')
+        stopped = total_schedulers - running
+        total_executions = sum(s.get('executions', 0) for s in states)
+        total_cost = sum(s.get('cost', 0.0) for s in states)
+        
+        print(f"\n{'='*60}")
+        print(f"📊 Aggregate Scheduler Statistics")
+        print(f"{'='*60}")
+        print(f"Total Schedulers:     {total_schedulers}")
+        print(f"  🟢 Running:         {running}")
+        print(f"  🔴 Stopped:         {stopped}")
+        print(f"\nTotal Executions:     {total_executions}")
+        print(f"Total Cost:           ${total_cost:.4f}")
+        
+        if total_executions > 0:
+            avg_cost = total_cost / total_executions
+            print(f"Avg Cost/Execution:   ${avg_cost:.4f}")
+        
+        print(f"{'='*60}\n")
+        
+        # Show per-scheduler breakdown
+        print("Per-Scheduler Breakdown:")
+        print(f"{'Name':<20} {'Status':<10} {'Executions':<12} {'Cost':<12}")
+        print("="*60)
+        
+        for state in sorted(states, key=lambda x: x.get('executions', 0), reverse=True):
+            name = state['name'][:20]
+            status = state.get('status', 'unknown')
+            executions = state.get('executions', 0)
+            cost = state.get('cost', 0.0)
+            
+            status_icon = "🟢" if status == "running" else "🔴"
+            print(f"{name:<20} {status_icon} {status:<8} {executions:<12} ${cost:<11.4f}")
+        
+        return 0
+    
+    @staticmethod
+    def _handle_stop_all(state_manager, daemon_manager) -> int:
+        """Handle 'schedule stop-all' command."""
+        states = state_manager.list_all()
+        
+        if not states:
+            print("No schedulers running.")
+            return 0
+        
+        print(f"\n🛑 Stopping all schedulers ({len(states)} total)...\n")
+        
+        stopped = 0
+        failed = 0
+        
+        for state in states:
+            name = state['name']
+            pid = state['pid']
+            
+            try:
+                if daemon_manager.stop_daemon(pid):
+                    state_manager.delete_state(name)
+                    print(f"✅ Stopped '{name}' (PID: {pid})")
+                    stopped += 1
+                else:
+                    print(f"❌ Failed to stop '{name}' (PID: {pid})")
+                    failed += 1
+            except Exception as e:
+                print(f"❌ Error stopping '{name}': {e}")
+                failed += 1
+        
+        print(f"\n📊 Summary: {stopped} stopped, {failed} failed")
+        return 0 if failed == 0 else 1
+    
+    @staticmethod
+    def _handle_stop(unknown_args, state_manager, daemon_manager) -> int:
+        """Handle 'schedule stop' command."""
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule stop <name>")
+            return 1
+        
+        name = unknown_args[0]
+        state = state_manager.load_state(name)
+        
+        if not state:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            print("\nList schedulers with: praisonai schedule list")
+            return 1
+        
+        pid = state.get('pid')
+        if not pid:
+            print(f"❌ Error: No PID found for scheduler '{name}'")
+            return 1
+        
+        print(f"🛑 Stopping scheduler '{name}' (PID: {pid})...")
+        
+        success = daemon_manager.stop_daemon(pid)
+        
+        if success:
+            state['status'] = 'stopped'
+            state_manager.save_state(name, state)
+            print(f"✅ Scheduler '{name}' stopped successfully")
+            return 0
+        else:
+            print(f"❌ Failed to stop scheduler '{name}'")
+            return 1
+    
+    @staticmethod
+    def _handle_logs(unknown_args, daemon_manager) -> int:
+        """Handle 'schedule logs' command."""
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule logs <name> [-f]")
+            return 1
+        
+        name = unknown_args[0]
+        follow = '-f' in unknown_args or '--follow' in unknown_args
+        
+        if follow:
+            print(f"📋 Following logs for '{name}' (Ctrl+C to stop)...")
+            import subprocess
+            log_file = daemon_manager.log_dir / f"{name}.log"
+            if not log_file.exists():
+                print(f"❌ Error: Log file not found for '{name}'")
+                return 1
+            
+            try:
+                subprocess.run(['tail', '-f', str(log_file)])
+            except KeyboardInterrupt:
+                print("\n✅ Stopped following logs")
+            return 0
+        else:
+            logs = daemon_manager.read_logs(name, lines=50)
+            if logs:
+                print(f"📋 Last 50 lines of logs for '{name}':\n")
+                print(logs)
+                return 0
+            else:
+                print(f"❌ Error: No logs found for '{name}'")
+                return 1
+    
+    @staticmethod
+    def _handle_restart(unknown_args, state_manager, daemon_manager) -> int:
+        """Handle 'schedule restart' command."""
+        from datetime import datetime
+        import time
+        
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule restart <name>")
+            return 1
+        
+        name = unknown_args[0]
+        state = state_manager.load_state(name)
+        
+        if not state:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            return 1
+        
+        print(f"🔄 Restarting scheduler '{name}'...")
+        
+        # Stop if running
+        pid = state.get('pid')
+        if pid and state_manager.is_process_alive(pid):
+            daemon_manager.stop_daemon(pid)
+            time.sleep(1)
+        
+        # Start again
+        new_pid = daemon_manager.start_scheduler_daemon(
+            name=name,
+            task=state['task'],
+            interval=state['interval'],
+            max_cost=state.get('max_cost'),
+            timeout=state.get('timeout'),
+            max_retries=state.get('max_retries', 3)
+        )
+        
+        state['pid'] = new_pid
+        state['status'] = 'running'
+        state['started_at'] = datetime.now().isoformat()
+        state_manager.save_state(name, state)
+        
+        print(f"✅ Scheduler '{name}' restarted (PID: {new_pid})")
+        return 0
+    
+    @staticmethod
+    def _handle_delete(unknown_args, state_manager) -> int:
+        """Handle 'schedule delete' command."""
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule delete <name>")
+            return 1
+        
+        name = unknown_args[0]
+        
+        if state_manager.delete_state(name):
+            print(f"✅ Scheduler '{name}' deleted from list")
+            return 0
+        else:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            return 1
+    
+    @staticmethod
+    def _handle_describe(unknown_args, state_manager, daemon_manager) -> int:
+        """Handle 'schedule describe' command."""
+        from datetime import datetime
+        
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule describe <name>")
+            return 1
+        
+        name = unknown_args[0]
+        state = state_manager.load_state(name)
+        
+        if not state:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            print("\nList schedulers with: praisonai schedule list")
+            return 1
+        
+        # Get process status
+        pid = state.get('pid', 0)
+        is_alive = state_manager.is_process_alive(pid)
+        status = "🟢 running" if is_alive else "🔴 stopped"
+        
+        # Calculate uptime
+        started_at = state.get('started_at')
+        uptime = "N/A"
+        if started_at:
+            try:
+                start_time = datetime.fromisoformat(started_at)
+                uptime_delta = datetime.now() - start_time
+                hours = int(uptime_delta.total_seconds() // 3600)
+                minutes = int((uptime_delta.total_seconds() % 3600) // 60)
+                uptime = f"{hours}h {minutes}m"
+            except:
+                pass
+        
+        # Display detailed info
+        print(f"\n{'='*60}")
+        print(f"📋 Scheduler Details: {name}")
+        print(f"{'='*60}")
+        print(f"Status:       {status}")
+        print(f"PID:          {pid}")
+        print(f"Uptime:       {uptime}")
+        print(f"Task:         {state.get('task', 'N/A')}")
+        print(f"Interval:     {state.get('interval', 'N/A')}")
+        print(f"Max Retries:  {state.get('max_retries', 'N/A')}")
+        
+        if state.get('timeout'):
+            print(f"Timeout:      {state['timeout']}s")
+        
+        if state.get('max_cost'):
+            print(f"Budget:       ${state['max_cost']}")
+        
+        print(f"Executions:   {state.get('executions', 0)}")
+        print(f"Total Cost:   ${state.get('cost', 0.0):.4f}")
+        print(f"Started:      {started_at or 'N/A'}")
+        
+        # Log file location
+        log_file = daemon_manager.log_dir / f"{name}.log"
+        print(f"Logs:         {log_file}")
+        
+        print(f"{'='*60}\n")
+        
+        return 0
+    
+    @staticmethod
+    def _handle_save(unknown_args, state_manager) -> int:
+        """Handle 'schedule save' command."""
+        import yaml
+        
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule save <name> [output.yaml]")
+            return 1
+        
+        name = unknown_args[0]
+        output_file = unknown_args[1] if len(unknown_args) > 1 else f"{name}.yaml"
+        
+        state = state_manager.load_state(name)
+        
+        if not state:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            return 1
+        
+        # Create YAML config from state
+        yaml_config = {
+            'framework': 'praisonai',
+            'agents': [{
+                'name': name,
+                'role': 'Task Executor',
+                'goal': state.get('task', ''),
+                'instructions': state.get('task', ''),
+                'verbose': True
+            }],
+            'task': state.get('task', ''),
+            'schedule': {
+                'interval': state.get('interval', 'hourly'),
+                'max_retries': state.get('max_retries', 3),
+                'run_immediately': True
+            }
+        }
+        
+        # Add optional fields
+        if state.get('timeout'):
+            yaml_config['schedule']['timeout'] = state['timeout']
+        
+        if state.get('max_cost'):
+            yaml_config['schedule']['max_cost'] = state['max_cost']
+        
+        # Write to file
+        try:
+            with open(output_file, 'w') as f:
+                yaml.dump(yaml_config, f, default_flow_style=False, sort_keys=False)
+            
+            print(f"✅ Configuration saved to: {output_file}")
+            print(f"\nRun with:")
+            print(f"  praisonai schedule {output_file}")
+            return 0
+        except Exception as e:
+            print(f"❌ Error saving configuration: {e}")
+            return 1
+    
+    @staticmethod
+    def handle_schedule_command(args, unknown_args, daemon_mode=False) -> int:
+        """
+        Handle the schedule command for running agents periodically.
+        
+        Supports two modes:
+        1. YAML mode: praisonai schedule agents.yaml
+        2. Prompt mode: praisonai schedule "Your task here" --interval hourly
+        
+        Args:
+            args: Parsed command-line arguments
+            unknown_args: Additional arguments after the command
+            
+        Returns:
+            Exit code (0 for success, 1 for error)
+        """
+        try:
+            from praisonai.scheduler import AgentScheduler
+            from praisonaiagents import Agent
+        except ImportError:
+            print("Error: praisonai.scheduler or praisonaiagents module not found")
+            print("Please ensure PraisonAI is properly installed")
+            print("pip install praisonai praisonaiagents")
+            return 1
+        
+        # Check if first arg is a YAML file or a prompt
+        first_arg = unknown_args[0] if unknown_args else None
+        is_yaml_mode = first_arg and (first_arg.endswith('.yaml') or first_arg.endswith('.yml'))
+        
+        # Get overrides from CLI
+        interval_override = getattr(args, 'schedule_interval', None) or 'hourly'
+        max_retries_override = getattr(args, 'schedule_max_retries', None) or 3
+        timeout_override = getattr(args, 'timeout', None)
+        max_cost_override = getattr(args, 'max_cost', None)
+        verbose = getattr(args, 'verbose', False)
+        
+        # Set up logging - only show logs if verbose
+        if verbose:
+            logging.basicConfig(
+                level=logging.INFO,
+                format='[%(asctime)s] %(levelname)s - %(message)s',
+                datefmt='%Y-%m-%d %H:%M:%S'
+            )
+        else:
+            # Suppress scheduler logs in non-verbose mode
+            logging.getLogger('praisonai.scheduler').setLevel(logging.WARNING)
+        
+        try:
+            if is_yaml_mode:
+                # YAML mode: Load from agents.yaml
+                yaml_path = first_arg
+                print(f"🤖 Loading agent configuration from: {yaml_path}")
+                scheduler = AgentScheduler.from_yaml(
+                    yaml_path=yaml_path,
+                    interval_override=interval_override,
+                    max_retries_override=max_retries_override,
+                    timeout_override=timeout_override,
+                    max_cost_override=max_cost_override
+                )
+            else:
+                # Prompt mode: Create agent from direct prompt
+                if not first_arg:
+                    print("❌ Error: Please provide either a YAML file or a task prompt")
+                    print("\nExamples:")
+                    print("  praisonai schedule agents.yaml")
+                    print('  praisonai schedule "Check news every hour" --interval hourly')
+                    return 1
+                
+                task = first_arg
+                
+                # Create agent
+                agent = Agent(
+                    name="Scheduled Agent",
+                    role="Task Executor",
+                    goal=task,
+                    instructions=task,
+                    verbose=True  # Enable verbose to see output in logs
+                )
+                
+                # Create scheduler
+                scheduler = AgentScheduler(
+                    agent=agent,
+                    task=task,
+                    timeout=timeout_override,
+                    max_cost=max_cost_override
+                )
+            
+            # Get configuration
+            interval = interval_override
+            max_retries = max_retries_override
+            
+            if is_yaml_mode:
+                schedule_config = scheduler._yaml_schedule_config
+                interval = interval_override or schedule_config.get('interval', 'hourly')
+                max_retries = max_retries_override or schedule_config.get('max_retries', 3)
+            
+            print(f"Schedule: {interval}")
+            print(f"Max Retries: {max_retries}")
+            if scheduler.timeout:
+                print(f"Timeout: {scheduler.timeout}s")
+            if scheduler.max_cost:
+                print(f"Budget: ${scheduler.max_cost}")
+            print(f"{'='*60}\n")
+            
+            # Start scheduler
+            if is_yaml_mode:
+                scheduler.start_from_yaml_config()
+            else:
+                scheduler.start(schedule_expr=interval, max_retries=max_retries, run_immediately=True)
+            
+            # If daemon mode, block to keep process alive
+            if daemon_mode:
+                import time
+                try:
+                    while scheduler.is_running:
+                        time.sleep(1)
+                except KeyboardInterrupt:
+                    scheduler.stop()
+            else:
+                # Foreground mode - wait for Ctrl+C
+                print("⏰ Starting scheduler... (Press Ctrl+C to stop)\n")
+                import time
+                try:
+                    while scheduler.is_running:
+                        time.sleep(1)
+                except KeyboardInterrupt:
+                    print("\n\n🛑 Stopping scheduler...")
+                    scheduler.stop()
+                    
+                    # Display final stats
+                    stats = scheduler.get_stats()
+                    print("\n📊 Final Statistics:")
+                    print(f"  Total Executions: {stats['total_executions']}")
+                    print(f"  Successful: {stats['successful_executions']}")
+                    print(f"  Failed: {stats['failed_executions']}")
+                    print(f"  Success Rate: {stats['success_rate']:.1f}%")
+                    print("\n✅ Agent stopped successfully\n")
+                
+            return 0
+            
+        except FileNotFoundError as e:
+            print(f"❌ Error: {e}")
+            print(f"\nMake sure {yaml_path} exists in the current directory.")
+            print("\nExample agents.yaml structure:")
+            print("""
+framework: praisonai
+
+agents:
+  - name: "AI News Monitor"
+    role: "News Analyst"
+    instructions: "Search and summarize AI news"
+    tools:
+      - search_tool
+
+task: "Search for latest AI news"
+
+schedule:
+  interval: "hourly"
+  max_retries: 3
+  run_immediately: true
+""")
+            return 1
+            
+        except ValueError as e:
+            print(f"❌ Configuration Error: {e}")
+            return 1
+            
+        except Exception as e:
+            print(f"❌ Error: {e}")
+            if verbose:
+                import traceback
+                traceback.print_exc()
+            return 1
+            
+    @staticmethod
+    def add_schedule_arguments(subparsers):
+        """
+        Add schedule subcommand and arguments to parser.
+        
+        Args:
+            subparsers: Subparsers object from argparse
+        """
+        schedule_parser = subparsers.add_parser(
+            'schedule',
+            help='Schedule an agent to run continuously at regular intervals'
+        )
+        
+        schedule_parser.add_argument(
+            'schedule_yaml',
+            nargs='?',
+            default='agents.yaml',
+            help='Path to agents.yaml file (default: agents.yaml)'
+        )
+        
+        schedule_parser.add_argument(
+            '--interval',
+            dest='schedule_interval',
+            type=str,
+            help='Override schedule interval (e.g., "hourly", "*/30m", "daily")'
+        )
+        
+        schedule_parser.add_argument(
+            '--max-retries',
+            dest='schedule_max_retries',
+            type=int,
+            help='Override maximum retry attempts (default: from YAML or 3)'
+        )
+        
+        schedule_parser.add_argument(
+            '--verbose', '-v',
+            action='store_true',
+            help='Enable verbose logging'
+        )
+        
+        schedule_parser.set_defaults(func=AgentSchedulerHandler.handle_schedule_command)
+
+
+def setup_scheduler_cli(subparsers):
+    """
+    Setup scheduler CLI commands.
+    
+    Args:
+        subparsers: Subparsers object from argparse
+    """
+    AgentSchedulerHandler.add_schedule_arguments(subparsers)
@@ -0,0 +1,729 @@
+"""
+Agent Scheduler CLI Feature.
+
+Provides CLI commands for scheduling agents to run 24/7 at regular intervals.
+"""
+
+import logging
+
+logger = logging.getLogger(__name__)
+
+
+class AgentSchedulerHandler:
+    """Handler for agent scheduler CLI commands."""
+    
+    @staticmethod
+    def handle_daemon_command(subcommand: str, args, unknown_args=None) -> int:
+        """
+        Handle daemon management commands (start, list, stop, logs, restart).
+        
+        Args:
+            subcommand: Command to execute (start, list, stop, logs, restart)
+            args: Parsed command-line arguments
+            unknown_args: Additional arguments
+            
+        Returns:
+            Exit code
+        """
+        from praisonai.scheduler.state_manager import SchedulerStateManager
+        from praisonai.scheduler.daemon_manager import DaemonManager
+        
+        state_manager = SchedulerStateManager()
+        daemon_manager = DaemonManager()
+        
+        if subcommand == "start":
+            return AgentSchedulerHandler._handle_start(args, unknown_args, state_manager, daemon_manager)
+        elif subcommand == "list":
+            return AgentSchedulerHandler._handle_list(state_manager)
+        elif subcommand == "stop":
+            return AgentSchedulerHandler._handle_stop(unknown_args, state_manager, daemon_manager)
+        elif subcommand == "logs":
+            return AgentSchedulerHandler._handle_logs(unknown_args, daemon_manager)
+        elif subcommand == "restart":
+            return AgentSchedulerHandler._handle_restart(unknown_args, state_manager, daemon_manager)
+        elif subcommand == "delete":
+            return AgentSchedulerHandler._handle_delete(unknown_args, state_manager)
+        elif subcommand == "describe":
+            return AgentSchedulerHandler._handle_describe(unknown_args, state_manager, daemon_manager)
+        elif subcommand == "save":
+            return AgentSchedulerHandler._handle_save(unknown_args, state_manager)
+        elif subcommand == "stop-all":
+            return AgentSchedulerHandler._handle_stop_all(state_manager, daemon_manager)
+        elif subcommand == "stats":
+            return AgentSchedulerHandler._handle_stats(state_manager, unknown_args, daemon_manager)
+        else:
+            print(f"❌ Unknown subcommand: {subcommand}")
+            print("\nAvailable commands:")
+            print("  start, list, stop, logs, restart, delete, describe, save, stop-all, stats")
+            return 1
+    
+    @staticmethod
+    def _handle_start(args, unknown_args, state_manager, daemon_manager) -> int:
+        """Handle 'schedule start' command."""
+        from datetime import datetime
+        
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name and task")
+            print("\nUsage:")
+            print('  praisonai schedule start <name> "Your task" --interval hourly')
+            print("\nExample:")
+            print('  praisonai schedule start news-checker "Check AI news" --interval hourly')
+            return 1
+        
+        # Parse arguments
+        name = unknown_args[0]
+        task = unknown_args[1] if len(unknown_args) > 1 else None
+        
+        if not task:
+            print("❌ Error: Please provide a task")
+            return 1
+        
+        # Get options
+        interval = getattr(args, 'schedule_interval', None) or 'hourly'
+        max_retries = getattr(args, 'schedule_max_retries', None) or 3
+        timeout = getattr(args, 'timeout', None)
+        max_cost = getattr(args, 'max_cost', None)
+        
+        # Check if name already exists
+        existing = state_manager.load_state(name)
+        if existing and daemon_manager.get_status(existing.get('pid', 0))['is_alive']:
+            print(f"❌ Error: Scheduler '{name}' is already running (PID: {existing['pid']})")
+            print(f"   Use 'praisonai schedule stop {name}' to stop it first")
+            return 1
+        
+        # Start daemon
+        print(f"🚀 Starting scheduler '{name}'...")
+        print(f"   Task: {task}")
+        print(f"   Interval: {interval}")
+        if timeout:
+            print(f"   Timeout: {timeout}s")
+        if max_cost:
+            print(f"   Budget: ${max_cost}")
+        
+        pid = daemon_manager.start_scheduler_daemon(
+            name=name,
+            task=task,
+            interval=interval,
+            max_cost=max_cost,
+            timeout=timeout,
+            max_retries=max_retries
+        )
+        
+        # Save state
+        state = {
+            "name": name,
+            "pid": pid,
+            "task": task,
+            "interval": interval,
+            "timeout": timeout,
+            "max_cost": max_cost,
+            "max_retries": max_retries,
+            "status": "running",
+            "started_at": datetime.now().isoformat(),
+            "executions": 0,
+            "cost": 0.0
+        }
+        state_manager.save_state(name, state)
+        
+        print(f"\n✅ Scheduler '{name}' started successfully!")
+        print(f"   PID: {pid}")
+        print(f"   Logs: ~/.praisonai/logs/{name}.log")
+        print(f"\nManage:")
+        print(f"   praisonai schedule list")
+        print(f"   praisonai schedule describe {name}")
+        print(f"   praisonai schedule logs {name} --follow")
+        print(f"   praisonai schedule stop {name}")
+        
+        return 0
+    
+    @staticmethod
+    def _handle_list(state_manager) -> int:
+        """Handle 'schedule list' command."""
+        states = state_manager.list_all()
+        
+        if not states:
+            print("No schedulers running.")
+            print("\nStart one with:")
+            print('  praisonai schedule start <name> "Your task" --interval hourly')
+            return 0
+        
+        # Clean up dead processes
+        state_manager.cleanup_dead_processes()
+        states = state_manager.list_all()
+        
+        # Display table
+        print(f"\n{'Name':<20} {'Status':<10} {'PID':<8} {'Interval':<12} {'Task':<40}")
+        print("=" * 90)
+        
+        for state in states:
+            name = state.get('name', 'unknown')[:20]
+            pid = state.get('pid', 0)
+            status = "running" if state_manager.is_process_alive(pid) else "stopped"
+            interval = state.get('interval', 'unknown')[:12]
+            task = state.get('task', '')[:40]
+            
+            status_icon = "🟢" if status == "running" else "🔴"
+            print(f"{name:<20} {status_icon} {status:<8} {pid:<8} {interval:<12} {task:<40}")
+        
+        print(f"\nTotal: {len(states)} scheduler(s)")
+        return 0
+    
+    @staticmethod
+    def _handle_stats(state_manager, unknown_args=None, daemon_manager=None) -> int:
+        """Handle 'schedule stats' command - show aggregate or individual statistics."""
+        # If a name is provided, show individual stats (alias for describe)
+        if unknown_args and len(unknown_args) > 0:
+            name = unknown_args[0]
+            return AgentSchedulerHandler._handle_describe([name], state_manager, None)
+        
+        # Otherwise show aggregate stats
+        states = state_manager.list_all()
+        
+        if not states:
+            print("No schedulers found.")
+            print("\nStart one with:")
+            print('  praisonai schedule start <name> "Your task" --interval hourly')
+            return 0
+        
+        # Calculate aggregate stats
+        total_schedulers = len(states)
+        running = sum(1 for s in states if s.get('status') == 'running')
+        stopped = total_schedulers - running
+        total_executions = sum(s.get('executions', 0) for s in states)
+        total_cost = sum(s.get('cost', 0.0) for s in states)
+        
+        print(f"\n{'='*60}")
+        print(f"📊 Aggregate Scheduler Statistics")
+        print(f"{'='*60}")
+        print(f"Total Schedulers:     {total_schedulers}")
+        print(f"  🟢 Running:         {running}")
+        print(f"  🔴 Stopped:         {stopped}")
+        print(f"\nTotal Executions:     {total_executions}")
+        print(f"Total Cost:           ${total_cost:.4f}")
+        
+        if total_executions > 0:
+            avg_cost = total_cost / total_executions
+            print(f"Avg Cost/Execution:   ${avg_cost:.4f}")
+        
+        print(f"{'='*60}\n")
+        
+        # Show per-scheduler breakdown
+        print("Per-Scheduler Breakdown:")
+        print(f"{'Name':<20} {'Status':<10} {'Executions':<12} {'Cost':<12}")
+        print("="*60)
+        
+        for state in sorted(states, key=lambda x: x.get('executions', 0), reverse=True):
+            name = state['name'][:20]
+            status = state.get('status', 'unknown')
+            executions = state.get('executions', 0)
+            cost = state.get('cost', 0.0)
+            
+            status_icon = "🟢" if status == "running" else "🔴"
+            print(f"{name:<20} {status_icon} {status:<8} {executions:<12} ${cost:<11.4f}")
+        
+        return 0
+    
+    @staticmethod
+    def _handle_stop_all(state_manager, daemon_manager) -> int:
+        """Handle 'schedule stop-all' command."""
+        states = state_manager.list_all()
+        
+        if not states:
+            print("No schedulers running.")
+            return 0
+        
+        print(f"\n🛑 Stopping all schedulers ({len(states)} total)...\n")
+        
+        stopped = 0
+        failed = 0
+        
+        for state in states:
+            name = state['name']
+            pid = state['pid']
+            
+            try:
+                if daemon_manager.stop_daemon(pid):
+                    state_manager.delete_state(name)
+                    print(f"✅ Stopped '{name}' (PID: {pid})")
+                    stopped += 1
+                else:
+                    print(f"❌ Failed to stop '{name}' (PID: {pid})")
+                    failed += 1
+            except Exception as e:
+                print(f"❌ Error stopping '{name}': {e}")
+                failed += 1
+        
+        print(f"\n📊 Summary: {stopped} stopped, {failed} failed")
+        return 0 if failed == 0 else 1
+    
+    @staticmethod
+    def _handle_stop(unknown_args, state_manager, daemon_manager) -> int:
+        """Handle 'schedule stop' command."""
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule stop <name>")
+            return 1
+        
+        name = unknown_args[0]
+        state = state_manager.load_state(name)
+        
+        if not state:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            print("\nList schedulers with: praisonai schedule list")
+            return 1
+        
+        pid = state.get('pid')
+        if not pid:
+            print(f"❌ Error: No PID found for scheduler '{name}'")
+            return 1
+        
+        print(f"🛑 Stopping scheduler '{name}' (PID: {pid})...")
+        
+        success = daemon_manager.stop_daemon(pid)
+        
+        if success:
+            state['status'] = 'stopped'
+            state_manager.save_state(name, state)
+            print(f"✅ Scheduler '{name}' stopped successfully")
+            return 0
+        else:
+            print(f"❌ Failed to stop scheduler '{name}'")
+            return 1
+    
+    @staticmethod
+    def _handle_logs(unknown_args, daemon_manager) -> int:
+        """Handle 'schedule logs' command."""
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule logs <name> [-f]")
+            return 1
+        
+        name = unknown_args[0]
+        follow = '-f' in unknown_args or '--follow' in unknown_args
+        
+        if follow:
+            print(f"📋 Following logs for '{name}' (Ctrl+C to stop)...")
+            import subprocess
+            log_file = daemon_manager.log_dir / f"{name}.log"
+            if not log_file.exists():
+                print(f"❌ Error: Log file not found for '{name}'")
+                return 1
+            
+            try:
+                subprocess.run(['tail', '-f', str(log_file)])
+            except KeyboardInterrupt:
+                print("\n✅ Stopped following logs")
+            return 0
+        else:
+            logs = daemon_manager.read_logs(name, lines=50)
+            if logs:
+                print(f"📋 Last 50 lines of logs for '{name}':\n")
+                print(logs)
+                return 0
+            else:
+                print(f"❌ Error: No logs found for '{name}'")
+                return 1
+    
+    @staticmethod
+    def _handle_restart(unknown_args, state_manager, daemon_manager) -> int:
+        """Handle 'schedule restart' command."""
+        from datetime import datetime
+        import time
+        
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule restart <name>")
+            return 1
+        
+        name = unknown_args[0]
+        state = state_manager.load_state(name)
+        
+        if not state:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            return 1
+        
+        print(f"🔄 Restarting scheduler '{name}'...")
+        
+        # Stop if running
+        pid = state.get('pid')
+        if pid and state_manager.is_process_alive(pid):
+            daemon_manager.stop_daemon(pid)
+            time.sleep(1)
+        
+        # Start again
+        new_pid = daemon_manager.start_scheduler_daemon(
+            name=name,
+            task=state['task'],
+            interval=state['interval'],
+            max_cost=state.get('max_cost'),
+            timeout=state.get('timeout'),
+            max_retries=state.get('max_retries', 3)
+        )
+        
+        state['pid'] = new_pid
+        state['status'] = 'running'
+        state['started_at'] = datetime.now().isoformat()
+        state_manager.save_state(name, state)
+        
+        print(f"✅ Scheduler '{name}' restarted (PID: {new_pid})")
+        return 0
+    
+    @staticmethod
+    def _handle_delete(unknown_args, state_manager) -> int:
+        """Handle 'schedule delete' command."""
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule delete <name>")
+            return 1
+        
+        name = unknown_args[0]
+        
+        if state_manager.delete_state(name):
+            print(f"✅ Scheduler '{name}' deleted from list")
+            return 0
+        else:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            return 1
+    
+    @staticmethod
+    def _handle_describe(unknown_args, state_manager, daemon_manager) -> int:
+        """Handle 'schedule describe' command."""
+        from datetime import datetime
+        
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule describe <name>")
+            return 1
+        
+        name = unknown_args[0]
+        state = state_manager.load_state(name)
+        
+        if not state:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            print("\nList schedulers with: praisonai schedule list")
+            return 1
+        
+        # Get process status
+        pid = state.get('pid', 0)
+        is_alive = state_manager.is_process_alive(pid)
+        status = "🟢 running" if is_alive else "🔴 stopped"
+        
+        # Calculate uptime
+        started_at = state.get('started_at')
+        uptime = "N/A"
+        if started_at:
+            try:
+                start_time = datetime.fromisoformat(started_at)
+                uptime_delta = datetime.now() - start_time
+                hours = int(uptime_delta.total_seconds() // 3600)
+                minutes = int((uptime_delta.total_seconds() % 3600) // 60)
+                uptime = f"{hours}h {minutes}m"
+            except:
+                pass
+        
+        # Display detailed info
+        print(f"\n{'='*60}")
+        print(f"📋 Scheduler Details: {name}")
+        print(f"{'='*60}")
+        print(f"Status:       {status}")
+        print(f"PID:          {pid}")
+        print(f"Uptime:       {uptime}")
+        print(f"Task:         {state.get('task', 'N/A')}")
+        print(f"Interval:     {state.get('interval', 'N/A')}")
+        print(f"Max Retries:  {state.get('max_retries', 'N/A')}")
+        
+        if state.get('timeout'):
+            print(f"Timeout:      {state['timeout']}s")
+        
+        if state.get('max_cost'):
+            print(f"Budget:       ${state['max_cost']}")
+        
+        print(f"Executions:   {state.get('executions', 0)}")
+        print(f"Total Cost:   ${state.get('cost', 0.0):.4f}")
+        print(f"Started:      {started_at or 'N/A'}")
+        
+        # Log file location
+        log_file = daemon_manager.log_dir / f"{name}.log"
+        print(f"Logs:         {log_file}")
+        
+        print(f"{'='*60}\n")
+        
+        return 0
+    
+    @staticmethod
+    def _handle_save(unknown_args, state_manager) -> int:
+        """Handle 'schedule save' command."""
+        import yaml
+        
+        if not unknown_args:
+            print("❌ Error: Please provide scheduler name")
+            print("\nUsage: praisonai schedule save <name> [output.yaml]")
+            return 1
+        
+        name = unknown_args[0]
+        output_file = unknown_args[1] if len(unknown_args) > 1 else f"{name}.yaml"
+        
+        state = state_manager.load_state(name)
+        
+        if not state:
+            print(f"❌ Error: Scheduler '{name}' not found")
+            return 1
+        
+        # Create YAML config from state
+        yaml_config = {
+            'framework': 'praisonai',
+            'agents': [{
+                'name': name,
+                'role': 'Task Executor',
+                'goal': state.get('task', ''),
+                'instructions': state.get('task', ''),
+                'verbose': True
+            }],
+            'task': state.get('task', ''),
+            'schedule': {
+                'interval': state.get('interval', 'hourly'),
+                'max_retries': state.get('max_retries', 3),
+                'run_immediately': True
+            }
+        }
+        
+        # Add optional fields
+        if state.get('timeout'):
+            yaml_config['schedule']['timeout'] = state['timeout']
+        
+        if state.get('max_cost'):
+            yaml_config['schedule']['max_cost'] = state['max_cost']
+        
+        # Write to file
+        try:
+            with open(output_file, 'w') as f:
+                yaml.dump(yaml_config, f, default_flow_style=False, sort_keys=False)
+            
+            print(f"✅ Configuration saved to: {output_file}")
+            print(f"\nRun with:")
+            print(f"  praisonai schedule {output_file}")
+            return 0
+        except Exception as e:
+            print(f"❌ Error saving configuration: {e}")
+            return 1
+    
+    @staticmethod
+    def handle_schedule_command(args, unknown_args=None) -> int:
+        """
+        Handle the schedule command for running agents periodically.
+        
+        Supports two modes:
+        1. YAML mode: praisonai schedule agents.yaml
+        2. Prompt mode: praisonai schedule "Your task here" --interval hourly
+        
+        Args:
+            args: Parsed command-line arguments
+            unknown_args: Additional arguments after the command
+            
+        Returns:
+            Exit code (0 for success, 1 for error)
+        """
+        try:
+            from praisonai.scheduler import AgentScheduler
+            from praisonaiagents import Agent
+        except ImportError:
+            print("Error: praisonai.scheduler or praisonaiagents module not found")
+            print("Please ensure PraisonAI is properly installed")
+            print("pip install praisonai praisonaiagents")
+            return 1
+        
+        # Check if first arg is a YAML file or a prompt
+        first_arg = unknown_args[0] if unknown_args else None
+        is_yaml_mode = first_arg and (first_arg.endswith('.yaml') or first_arg.endswith('.yml'))
+        
+        # Get overrides from CLI
+        interval_override = getattr(args, 'schedule_interval', None) or 'hourly'
+        max_retries_override = getattr(args, 'schedule_max_retries', None) or 3
+        timeout_override = getattr(args, 'timeout', None)
+        max_cost_override = getattr(args, 'max_cost', None)
+        verbose = getattr(args, 'verbose', False)
+        
+        # Set up logging - only show logs if verbose
+        if verbose:
+            logging.basicConfig(
+                level=logging.INFO,
+                format='[%(asctime)s] %(levelname)s - %(message)s',
+                datefmt='%Y-%m-%d %H:%M:%S'
+            )
+        else:
+            # Suppress scheduler logs in non-verbose mode
+            logging.getLogger('praisonai.scheduler').setLevel(logging.WARNING)
+        
+        try:
+            if is_yaml_mode:
+                # YAML mode: Load from agents.yaml
+                yaml_path = first_arg
+                print(f"🤖 Loading agent configuration from: {yaml_path}")
+                scheduler = AgentScheduler.from_yaml(
+                    yaml_path=yaml_path,
+                    interval_override=interval_override,
+                    max_retries_override=max_retries_override,
+                    timeout_override=timeout_override,
+                    max_cost_override=max_cost_override
+                )
+            else:
+                # Prompt mode: Create agent from direct prompt
+                if not first_arg:
+                    print("❌ Error: Please provide either a YAML file or a task prompt")
+                    print("\nExamples:")
+                    print("  praisonai schedule agents.yaml")
+                    print('  praisonai schedule "Check news every hour" --interval hourly')
+                    return 1
+                
+                task = first_arg
+                
+                # Create agent
+                agent = Agent(
+                    name="Scheduled Agent",
+                    role="Task Executor",
+                    goal=task,
+                    instructions=task,
+                    verbose=False
+                )
+                
+                # Create scheduler
+                scheduler = AgentScheduler(
+                    agent=agent,
+                    task=task,
+                    timeout=timeout_override,
+                    max_cost=max_cost_override
+                )
+            
+            # Get configuration
+            interval = interval_override
+            max_retries = max_retries_override
+            
+            if is_yaml_mode:
+                schedule_config = scheduler._yaml_schedule_config
+                interval = interval_override or schedule_config.get('interval', 'hourly')
+                max_retries = max_retries_override or schedule_config.get('max_retries', 3)
+            
+            print(f"Schedule: {interval}")
+            print(f"Max Retries: {max_retries}")
+            if scheduler.timeout:
+                print(f"Timeout: {scheduler.timeout}s")
+            if scheduler.max_cost:
+                print(f"Budget: ${scheduler.max_cost}")
+            print(f"{'='*60}\n")
+            
+            # Start scheduler
+            print("⏰ Starting scheduler... (Press Ctrl+C to stop)\n")
+            if is_yaml_mode:
+                scheduler.start_from_yaml_config()
+            else:
+                scheduler.start(
+                    schedule_expr=interval,
+                    max_retries=max_retries,
+                    run_immediately=True
+                )
+            
+            # Keep running until interrupted
+            try:
+                while scheduler.is_running:
+                    import time
+                    time.sleep(1)
+            except KeyboardInterrupt:
+                print("\n\n🛑 Stopping scheduler...")
+                scheduler.stop()
+                
+                # Display final stats
+                stats = scheduler.get_stats()
+                print("\n📊 Final Statistics:")
+                print(f"  Total Executions: {stats['total_executions']}")
+                print(f"  Successful: {stats['successful_executions']}")
+                print(f"  Failed: {stats['failed_executions']}")
+                print(f"  Success Rate: {stats['success_rate']:.1f}%")
+                print("\n✅ Agent stopped successfully\n")
+                
+            return 0
+            
+        except FileNotFoundError as e:
+            print(f"❌ Error: {e}")
+            print(f"\nMake sure {yaml_path} exists in the current directory.")
+            print("\nExample agents.yaml structure:")
+            print("""
+framework: praisonai
+
+agents:
+  - name: "AI News Monitor"
+    role: "News Analyst"
+    instructions: "Search and summarize AI news"
+    tools:
+      - search_tool
+
+task: "Search for latest AI news"
+
+schedule:
+  interval: "hourly"
+  max_retries: 3
+  run_immediately: true
+""")
+            return 1
+            
+        except ValueError as e:
+            print(f"❌ Configuration Error: {e}")
+            return 1
+            
+        except Exception as e:
+            print(f"❌ Error: {e}")
+            if verbose:
+                import traceback
+                traceback.print_exc()
+            return 1
+            
+    @staticmethod
+    def add_schedule_arguments(subparsers):
+        """
+        Add schedule subcommand and arguments to parser.
+        
+        Args:
+            subparsers: Subparsers object from argparse
+        """
+        schedule_parser = subparsers.add_parser(
+            'schedule',
+            help='Schedule an agent to run continuously at regular intervals'
+        )
+        
+        schedule_parser.add_argument(
+            'schedule_yaml',
+            nargs='?',
+            default='agents.yaml',
+            help='Path to agents.yaml file (default: agents.yaml)'
+        )
+        
+        schedule_parser.add_argument(
+            '--interval',
+            dest='schedule_interval',
+            type=str,
+            help='Override schedule interval (e.g., "hourly", "*/30m", "daily")'
+        )
+        
+        schedule_parser.add_argument(
+            '--max-retries',
+            dest='schedule_max_retries',
+            type=int,
+            help='Override maximum retry attempts (default: from YAML or 3)'
+        )
+        
+        schedule_parser.add_argument(
+            '--verbose', '-v',
+            action='store_true',
+            help='Enable verbose logging'
+        )
+        
+        schedule_parser.set_defaults(func=AgentSchedulerHandler.handle_schedule_command)
+
+
+def setup_scheduler_cli(subparsers):
+    """
+    Setup scheduler CLI commands.
+    
+    Args:
+        subparsers: Subparsers object from argparse
+    """
+    AgentSchedulerHandler.add_schedule_arguments(subparsers)
@@ -0,0 +1,416 @@
+"""
+@ Mention Autocomplete for PraisonAI CLI.
+
+Provides file/directory autocomplete when user types @.
+Uses only stdlib imports for zero performance impact.
+
+Features:
+- Detect @ trigger in input
+- Fuzzy search files/directories
+- Cache results for performance
+- Integrate with prompt_toolkit
+"""
+
+import os
+import time
+import fnmatch
+from dataclasses import dataclass
+from typing import List, Optional, Dict, Tuple
+
+
+# ============================================================================
+# Data Classes
+# ============================================================================
+
+@dataclass
+class AtMentionContext:
+    """Context for an @ mention in user input."""
+    is_active: bool
+    query: str
+    start_pos: int
+    mention_type: str = "file"  # file, directory, web, etc.
+
+
+@dataclass
+class FileSuggestion:
+    """A file or directory suggestion."""
+    path: str
+    file_type: str  # "file" or "directory"
+    score: int = 100
+    display: str = ""
+    
+    def __post_init__(self):
+        if not self.display:
+            self.display = self.path
+
+
+# ============================================================================
+# @ Mention Detection
+# ============================================================================
+
+def detect_at_mention(text: str, cursor_pos: int) -> Optional[AtMentionContext]:
+    """
+    Detect if cursor is within an @ mention context.
+    
+    Scans backward from cursor to find @ symbol.
+    Returns None if not in @ context.
+    
+    Args:
+        text: Full input text
+        cursor_pos: Current cursor position
+        
+    Returns:
+        AtMentionContext if in @ mention, None otherwise
+    """
+    if cursor_pos <= 0 or cursor_pos > len(text):
+        return None
+    
+    # Scan backward from cursor to find @
+    at_pos = -1
+    for i in range(cursor_pos - 1, -1, -1):
+        char = text[i]
+        
+        # Found @
+        if char == '@':
+            at_pos = i
+            break
+        
+        # Space/newline breaks @ context
+        if char in ' \t\n\r':
+            return None
+    
+    if at_pos == -1:
+        return None
+    
+    # Extract query between @ and cursor
+    query = text[at_pos + 1:cursor_pos]
+    
+    return AtMentionContext(
+        is_active=True,
+        query=query,
+        start_pos=at_pos
+    )
+
+
+# ============================================================================
+# File Search Service
+# ============================================================================
+
+class FileSearchService:
+    """
+    Service for searching files in a directory.
+    
+    Features:
+    - Walk directory tree
+    - Fuzzy match file paths
+    - Cache results with TTL
+    - Respect .gitignore (basic)
+    """
+    
+    def __init__(
+        self,
+        root_dir: str,
+        cache_ttl: int = 30,
+        max_depth: int = 5,
+        ignore_patterns: Optional[List[str]] = None
+    ):
+        self._root = os.path.abspath(root_dir)
+        self._cache_ttl = cache_ttl
+        self._max_depth = max_depth
+        self._ignore_patterns = ignore_patterns or [
+            '.git', '__pycache__', 'node_modules', '.venv', 'venv',
+            '*.pyc', '*.pyo', '.DS_Store', '*.egg-info'
+        ]
+        
+        # Cache: query -> (results, timestamp)
+        self._cache: Dict[str, Tuple[List[FileSuggestion], float]] = {}
+        
+        # File list cache
+        self._file_list: Optional[List[str]] = None
+        self._file_list_time: float = 0
+    
+    def _should_ignore(self, name: str) -> bool:
+        """Check if file/dir should be ignored."""
+        for pattern in self._ignore_patterns:
+            if fnmatch.fnmatch(name, pattern):
+                return True
+        return False
+    
+    def _get_file_list(self) -> List[str]:
+        """Get list of all files (cached)."""
+        now = time.time()
+        
+        # Return cached if fresh
+        if self._file_list is not None and (now - self._file_list_time) < self._cache_ttl:
+            return self._file_list
+        
+        files = []
+        
+        try:
+            for root, dirs, filenames in os.walk(self._root):
+                # Calculate depth
+                rel_root = os.path.relpath(root, self._root)
+                depth = 0 if rel_root == '.' else rel_root.count(os.sep) + 1
+                
+                if depth > self._max_depth:
+                    dirs.clear()  # Don't descend further
+                    continue
+                
+                # Filter ignored directories
+                dirs[:] = [d for d in dirs if not self._should_ignore(d)]
+                
+                # Add directories
+                for d in dirs:
+                    rel_path = os.path.relpath(os.path.join(root, d), self._root)
+                    files.append(rel_path + os.sep)
+                
+                # Add files
+                for f in filenames:
+                    if not self._should_ignore(f):
+                        rel_path = os.path.relpath(os.path.join(root, f), self._root)
+                        files.append(rel_path)
+        except OSError:
+            pass
+        
+        self._file_list = files
+        self._file_list_time = now
+        return files
+    
+    def _fuzzy_match(self, text: str, query: str) -> int:
+        """
+        Simple fuzzy match scoring.
+        
+        Returns score 0-100, higher is better.
+        """
+        if not query:
+            return 50  # Default score for empty query
+        
+        text_lower = text.lower()
+        query_lower = query.lower()
+        
+        # Exact match
+        if text_lower == query_lower:
+            return 100
+        
+        # Starts with
+        if text_lower.startswith(query_lower):
+            return 90
+        
+        # Contains
+        if query_lower in text_lower:
+            return 80
+        
+        # Fuzzy: all query chars appear in order
+        text_idx = 0
+        query_idx = 0
+        matches = 0
+        
+        while text_idx < len(text_lower) and query_idx < len(query_lower):
+            if text_lower[text_idx] == query_lower[query_idx]:
+                matches += 1
+                query_idx += 1
+            text_idx += 1
+        
+        if query_idx == len(query_lower):
+            # All query chars found
+            return 50 + int(30 * matches / len(text))
+        
+        return 0
+    
+    def search(self, query: str, max_results: int = 20) -> List[FileSuggestion]:
+        """
+        Search for files matching query.
+        
+        Args:
+            query: Search query (fuzzy matched)
+            max_results: Maximum results to return
+            
+        Returns:
+            List of FileSuggestion sorted by score
+        """
+        # Check cache
+        cache_key = f"{query}:{max_results}"
+        now = time.time()
+        
+        if cache_key in self._cache:
+            results, cached_time = self._cache[cache_key]
+            if (now - cached_time) < self._cache_ttl:
+                return results
+        
+        # Get all files
+        all_files = self._get_file_list()
+        
+        # Score and filter
+        scored: List[Tuple[int, str]] = []
+        for path in all_files:
+            score = self._fuzzy_match(path, query)
+            if score > 0:
+                scored.append((score, path))
+        
+        # Sort by score descending
+        scored.sort(key=lambda x: (-x[0], x[1]))
+        
+        # Build results
+        results = []
+        for score, path in scored[:max_results]:
+            file_type = "directory" if path.endswith(os.sep) else "file"
+            results.append(FileSuggestion(
+                path=path,
+                file_type=file_type,
+                score=score
+            ))
+        
+        # Cache results
+        self._cache[cache_key] = (results, now)
+        
+        return results
+    
+    def clear_cache(self) -> None:
+        """Clear all caches."""
+        self._cache.clear()
+        self._file_list = None
+
+
+# ============================================================================
+# Completers for prompt_toolkit
+# ============================================================================
+
+# Lazy import to avoid performance impact
+_Completer = None
+_Completion = None
+
+def _get_prompt_toolkit():
+    """Lazy import prompt_toolkit."""
+    global _Completer, _Completion
+    if _Completer is None:
+        from prompt_toolkit.completion import Completer, Completion
+        _Completer = Completer
+        _Completion = Completion
+    return _Completer, _Completion
+
+
+class AtMentionCompleter:
+    """
+    Completer for @ mentions.
+    
+    Shows file suggestions when user types @.
+    Implements prompt_toolkit Completer interface.
+    """
+    
+    def __init__(self, root_dir: str):
+        self._file_service = FileSearchService(root_dir)
+    
+    def get_completions(self, document, complete_event):
+        """Get completions for @ mentions."""
+        _, Completion = _get_prompt_toolkit()
+        
+        text = document.text_before_cursor
+        cursor_pos = len(text)
+        
+        # Detect @ context
+        context = detect_at_mention(text, cursor_pos)
+        if context is None or not context.is_active:
+            return
+        
+        # Search files
+        results = self._file_service.search(context.query, max_results=15)
+        
+        # Yield completions
+        for suggestion in results:
+            # Icon based on type
+            icon = "📁 " if suggestion.file_type == "directory" else "📄 "
+            display = f"{icon}{suggestion.path}"
+            
+            # Calculate replacement position
+            # Replace from @ to cursor
+            start_position = -(cursor_pos - context.start_pos)
+            
+            yield Completion(
+                text=f"@{suggestion.path}",
+                start_position=start_position,
+                display=display
+            )
+
+
+def create_combined_completer(commands: List[str], root_dir: str):
+    """
+    Factory function to create a CombinedCompleter.
+    
+    Returns a proper Completer subclass that handles both / and @.
+    """
+    Completer, Completion = _get_prompt_toolkit()
+    file_service = FileSearchService(root_dir)
+    
+    class CombinedCompleter(Completer):
+        """
+        Combined completer for both / commands and @ mentions.
+        
+        Properly inherits from prompt_toolkit Completer.
+        """
+        
+        def get_completions(self, document, complete_event):
+            """Get completions for / or @."""
+            text = document.text_before_cursor.lstrip()
+            
+            # Check for slash command
+            if text.startswith('/'):
+                cmd_text = text[1:].lower()
+                for cmd in commands:
+                    if cmd.lower().startswith(cmd_text):
+                        yield Completion(
+                            f'/{cmd}',
+                            start_position=-len(text),
+                            display=f'/{cmd}'
+                        )
+                return
+            
+            # Check for @ mention
+            full_text = document.text_before_cursor
+            cursor_pos = len(full_text)
+            
+            context = detect_at_mention(full_text, cursor_pos)
+            if context is None or not context.is_active:
+                return
+            
+            # Search files
+            results = file_service.search(context.query, max_results=15)
+            
+            # Yield completions
+            for suggestion in results:
+                icon = "📁 " if suggestion.file_type == "directory" else "📄 "
+                display = f"{icon}{suggestion.path}"
+                start_position = -(cursor_pos - context.start_pos)
+                
+                yield Completion(
+                    text=f"@{suggestion.path}",
+                    start_position=start_position,
+                    display=display
+                )
+    
+    return CombinedCompleter()
+
+
+# Backward compatible alias
+class CombinedCompleter:
+    """
+    Wrapper class for backward compatibility.
+    
+    Use create_combined_completer() for proper Completer subclass.
+    """
+    
+    def __new__(cls, commands: List[str], root_dir: str):
+        return create_combined_completer(commands, root_dir)
+
+
+# ============================================================================
+# Exports
+# ============================================================================
+
+__all__ = [
+    'AtMentionContext',
+    'FileSuggestion',
+    'detect_at_mention',
+    'FileSearchService',
+    'AtMentionCompleter',
+    'CombinedCompleter',
+]
@@ -0,0 +1,490 @@
+"""
+Autonomy Modes for PraisonAI CLI.
+
+Inspired by Codex CLI's approval modes and Gemini CLI's approval settings.
+Provides configurable autonomy levels: suggest, auto-edit, full-auto.
+
+Architecture:
+- AutonomyMode: Enum for different autonomy levels
+- AutonomyPolicy: Policy configuration for each mode
+- AutonomyManager: Manages mode transitions and approvals
+"""
+
+from dataclasses import dataclass, field
+from typing import Any, Callable, Dict, Optional, Set
+from enum import Enum
+import logging
+
+logger = logging.getLogger(__name__)
+
+
+class AutonomyMode(Enum):
+    """
+    Autonomy levels for agent execution.
+    
+    Inspired by Codex CLI's approval modes:
+    - SUGGEST: Read-only, requires approval for all changes
+    - AUTO_EDIT: Auto-approve file edits, require approval for commands
+    - FULL_AUTO: Auto-approve everything (YOLO mode)
+    """
+    SUGGEST = "suggest"          # Default: ask for approval on everything
+    AUTO_EDIT = "auto_edit"      # Auto-approve file edits only
+    FULL_AUTO = "full_auto"      # Auto-approve everything (dangerous)
+    
+    @classmethod
+    def from_string(cls, value: str) -> "AutonomyMode":
+        """Parse mode from string."""
+        value = value.lower().strip().replace("-", "_")
+        for mode in cls:
+            if mode.value == value:
+                return mode
+        raise ValueError(f"Unknown autonomy mode: {value}")
+
+
+class ActionType(Enum):
+    """Types of actions that require approval."""
+    FILE_READ = "file_read"
+    FILE_WRITE = "file_write"
+    FILE_DELETE = "file_delete"
+    SHELL_COMMAND = "shell_command"
+    NETWORK_REQUEST = "network_request"
+    CODE_EXECUTION = "code_execution"
+    GIT_OPERATION = "git_operation"
+    TOOL_CALL = "tool_call"
+
+
+@dataclass
+class ActionRequest:
+    """
+    Request for an action that may require approval.
+    
+    Attributes:
+        action_type: Type of action
+        description: Human-readable description
+        details: Additional details (file path, command, etc.)
+        risk_level: Estimated risk (low, medium, high)
+        reversible: Whether the action can be undone
+    """
+    action_type: ActionType
+    description: str
+    details: Dict[str, Any] = field(default_factory=dict)
+    risk_level: str = "medium"
+    reversible: bool = True
+    
+    def __str__(self) -> str:
+        return f"{self.action_type.value}: {self.description}"
+
+
+@dataclass
+class ApprovalResult:
+    """Result of an approval request."""
+    approved: bool
+    reason: Optional[str] = None
+    modified_action: Optional[ActionRequest] = None
+    remember_choice: bool = False
+
+
+@dataclass
+class AutonomyPolicy:
+    """
+    Policy configuration for autonomy mode.
+    
+    Defines which actions are auto-approved, require approval,
+    or are blocked entirely.
+    """
+    mode: AutonomyMode
+    
+    # Actions that are auto-approved
+    auto_approve: Set[ActionType] = field(default_factory=set)
+    
+    # Actions that always require approval
+    require_approval: Set[ActionType] = field(default_factory=set)
+    
+    # Actions that are blocked entirely
+    blocked: Set[ActionType] = field(default_factory=set)
+    
+    # Trusted paths for file operations
+    trusted_paths: Set[str] = field(default_factory=set)
+    
+    # Trusted commands (patterns)
+    trusted_commands: Set[str] = field(default_factory=set)
+    
+    # Maximum auto-approved actions before requiring confirmation
+    max_auto_actions: int = 10
+    
+    # Whether to show what would be done in suggest mode
+    show_preview: bool = True
+    
+    @classmethod
+    def for_mode(cls, mode: AutonomyMode) -> "AutonomyPolicy":
+        """Create default policy for a given mode."""
+        if mode == AutonomyMode.SUGGEST:
+            return cls(
+                mode=mode,
+                auto_approve={ActionType.FILE_READ},
+                require_approval={
+                    ActionType.FILE_WRITE,
+                    ActionType.FILE_DELETE,
+                    ActionType.SHELL_COMMAND,
+                    ActionType.CODE_EXECUTION,
+                    ActionType.GIT_OPERATION,
+                    ActionType.NETWORK_REQUEST,
+                    ActionType.TOOL_CALL,
+                },
+                blocked=set(),
+                show_preview=True
+            )
+        elif mode == AutonomyMode.AUTO_EDIT:
+            return cls(
+                mode=mode,
+                auto_approve={
+                    ActionType.FILE_READ,
+                    ActionType.FILE_WRITE,
+                    ActionType.TOOL_CALL,
+                },
+                require_approval={
+                    ActionType.FILE_DELETE,
+                    ActionType.SHELL_COMMAND,
+                    ActionType.CODE_EXECUTION,
+                    ActionType.GIT_OPERATION,
+                    ActionType.NETWORK_REQUEST,
+                },
+                blocked=set(),
+                show_preview=True
+            )
+        elif mode == AutonomyMode.FULL_AUTO:
+            return cls(
+                mode=mode,
+                auto_approve={
+                    ActionType.FILE_READ,
+                    ActionType.FILE_WRITE,
+                    ActionType.FILE_DELETE,
+                    ActionType.SHELL_COMMAND,
+                    ActionType.CODE_EXECUTION,
+                    ActionType.GIT_OPERATION,
+                    ActionType.NETWORK_REQUEST,
+                    ActionType.TOOL_CALL,
+                },
+                require_approval=set(),
+                blocked=set(),
+                show_preview=False,
+                max_auto_actions=100
+            )
+        else:
+            raise ValueError(f"Unknown mode: {mode}")
+
+
+class AutonomyManager:
+    """
+    Manages autonomy mode and approval flow.
+    
+    Handles:
+    - Mode transitions
+    - Approval requests
+    - Action tracking
+    - Policy enforcement
+    """
+    
+    def __init__(
+        self,
+        mode: AutonomyMode = AutonomyMode.SUGGEST,
+        policy: Optional[AutonomyPolicy] = None,
+        approval_callback: Optional[Callable[[ActionRequest], ApprovalResult]] = None,
+        verbose: bool = False
+    ):
+        self.mode = mode
+        self.policy = policy or AutonomyPolicy.for_mode(mode)
+        self.approval_callback = approval_callback or self._default_approval
+        self.verbose = verbose
+        
+        # Tracking
+        self._action_count = 0
+        self._auto_approved_count = 0
+        self._denied_count = 0
+        self._remembered_approvals: Dict[str, bool] = {}
+    
+    def set_mode(self, mode: AutonomyMode) -> None:
+        """Change the autonomy mode."""
+        old_mode = self.mode
+        self.mode = mode
+        self.policy = AutonomyPolicy.for_mode(mode)
+        
+        if self.verbose:
+            logger.info(f"Autonomy mode changed: {old_mode.value} -> {mode.value}")
+    
+    def request_approval(self, action: ActionRequest) -> ApprovalResult:
+        """
+        Request approval for an action.
+        
+        Returns:
+            ApprovalResult with approval decision
+        """
+        self._action_count += 1
+        
+        # Check if blocked
+        if action.action_type in self.policy.blocked:
+            self._denied_count += 1
+            return ApprovalResult(
+                approved=False,
+                reason=f"Action type '{action.action_type.value}' is blocked by policy"
+            )
+        
+        # Check remembered approvals
+        action_key = self._get_action_key(action)
+        if action_key in self._remembered_approvals:
+            approved = self._remembered_approvals[action_key]
+            if approved:
+                self._auto_approved_count += 1
+            else:
+                self._denied_count += 1
+            return ApprovalResult(approved=approved, reason="Remembered choice")
+        
+        # Check if auto-approved
+        if action.action_type in self.policy.auto_approve:
+            # Check max auto actions
+            if self._auto_approved_count >= self.policy.max_auto_actions:
+                # Force approval after max auto actions
+                return self._request_user_approval(action)
+            
+            self._auto_approved_count += 1
+            return ApprovalResult(approved=True, reason="Auto-approved by policy")
+        
+        # Check if requires approval
+        if action.action_type in self.policy.require_approval:
+            return self._request_user_approval(action)
+        
+        # Default: require approval for unknown action types
+        return self._request_user_approval(action)
+    
+    def _request_user_approval(self, action: ActionRequest) -> ApprovalResult:
+        """Request approval from user via callback."""
+        result = self.approval_callback(action)
+        
+        if result.remember_choice:
+            action_key = self._get_action_key(action)
+            self._remembered_approvals[action_key] = result.approved
+        
+        if result.approved:
+            self._auto_approved_count += 1
+        else:
+            self._denied_count += 1
+        
+        return result
+    
+    def _get_action_key(self, action: ActionRequest) -> str:
+        """Generate a key for remembering approval decisions."""
+        # Use action type and relevant details
+        key_parts = [action.action_type.value]
+        
+        if "path" in action.details:
+            key_parts.append(action.details["path"])
+        if "command" in action.details:
+            key_parts.append(action.details["command"])
+        
+        return ":".join(key_parts)
+    
+    def _default_approval(self, action: ActionRequest) -> ApprovalResult:
+        """Default approval callback using console input."""
+        from rich.console import Console
+        from rich.panel import Panel
+        from rich.prompt import Confirm
+        
+        console = Console()
+        
+        # Build action description
+        risk_color = {
+            "low": "green",
+            "medium": "yellow",
+            "high": "red"
+        }.get(action.risk_level, "white")
+        
+        details_str = "\n".join(
+            f"  {k}: {v}" for k, v in action.details.items()
+        ) if action.details else "  (no details)"
+        
+        panel_content = f"""
+[bold]{action.description}[/bold]
+
+Type: {action.action_type.value}
+Risk: [{risk_color}]{action.risk_level}[/{risk_color}]
+Reversible: {"Yes" if action.reversible else "No"}
+
+Details:
+{details_str}
+"""
+        
+        console.print(Panel(
+            panel_content,
+            title="🔒 Approval Required",
+            border_style="yellow"
+        ))
+        
+        approved = Confirm.ask("Approve this action?", default=False)
+        remember = False
+        
+        if approved:
+            remember = Confirm.ask("Remember this choice for similar actions?", default=False)
+        
+        return ApprovalResult(
+            approved=approved,
+            remember_choice=remember
+        )
+    
+    def get_stats(self) -> Dict[str, int]:
+        """Get approval statistics."""
+        return {
+            "total_actions": self._action_count,
+            "auto_approved": self._auto_approved_count,
+            "denied": self._denied_count,
+            "remembered": len(self._remembered_approvals)
+        }
+    
+    def reset_stats(self) -> None:
+        """Reset approval statistics."""
+        self._action_count = 0
+        self._auto_approved_count = 0
+        self._denied_count = 0
+    
+    def clear_remembered(self) -> None:
+        """Clear remembered approval decisions."""
+        self._remembered_approvals.clear()
+
+
+# ============================================================================
+# CLI Integration Handler
+# ============================================================================
+
+class AutonomyModeHandler:
+    """
+    Handler for integrating autonomy modes with PraisonAI CLI.
+    """
+    
+    def __init__(self, verbose: bool = False):
+        self.verbose = verbose
+        self._manager: Optional[AutonomyManager] = None
+    
+    @property
+    def feature_name(self) -> str:
+        return "autonomy_mode"
+    
+    def initialize(
+        self,
+        mode: str = "suggest",
+        approval_callback: Optional[Callable[[ActionRequest], ApprovalResult]] = None
+    ) -> AutonomyManager:
+        """
+        Initialize the autonomy manager.
+        
+        Args:
+            mode: Autonomy mode string (suggest, auto_edit, full_auto)
+            approval_callback: Optional custom approval callback
+            
+        Returns:
+            Configured AutonomyManager
+        """
+        try:
+            autonomy_mode = AutonomyMode.from_string(mode)
+        except ValueError:
+            logger.warning(f"Unknown mode '{mode}', defaulting to 'suggest'")
+            autonomy_mode = AutonomyMode.SUGGEST
+        
+        self._manager = AutonomyManager(
+            mode=autonomy_mode,
+            approval_callback=approval_callback,
+            verbose=self.verbose
+        )
+        
+        if self.verbose:
+            from rich import print as rprint
+            rprint(f"[cyan]Autonomy mode: {autonomy_mode.value}[/cyan]")
+        
+        return self._manager
+    
+    def get_manager(self) -> Optional[AutonomyManager]:
+        """Get the current autonomy manager."""
+        return self._manager
+    
+    def request_approval(self, action: ActionRequest) -> ApprovalResult:
+        """Request approval for an action."""
+        if not self._manager:
+            self._manager = self.initialize()
+        
+        return self._manager.request_approval(action)
+    
+    def set_mode(self, mode: str) -> None:
+        """Change the autonomy mode."""
+        if not self._manager:
+            self.initialize(mode)
+        else:
+            try:
+                autonomy_mode = AutonomyMode.from_string(mode)
+                self._manager.set_mode(autonomy_mode)
+            except ValueError as e:
+                logger.error(f"Failed to set mode: {e}")
+    
+    def get_mode(self) -> str:
+        """Get the current autonomy mode."""
+        if self._manager:
+            return self._manager.mode.value
+        return AutonomyMode.SUGGEST.value
+
+
+# ============================================================================
+# Convenience Functions
+# ============================================================================
+
+def create_file_write_action(path: str, content_preview: str = "") -> ActionRequest:
+    """Create an action request for file write."""
+    return ActionRequest(
+        action_type=ActionType.FILE_WRITE,
+        description=f"Write to file: {path}",
+        details={
+            "path": path,
+            "preview": content_preview[:200] if content_preview else "(no preview)"
+        },
+        risk_level="medium",
+        reversible=True
+    )
+
+
+def create_shell_command_action(command: str, cwd: str = ".") -> ActionRequest:
+    """Create an action request for shell command."""
+    # Assess risk based on command
+    dangerous_patterns = ["rm ", "sudo ", "chmod ", "chown ", "> /", "| sh", "curl |"]
+    risk = "high" if any(p in command for p in dangerous_patterns) else "medium"
+    
+    return ActionRequest(
+        action_type=ActionType.SHELL_COMMAND,
+        description=f"Execute command: {command}",
+        details={
+            "command": command,
+            "cwd": cwd
+        },
+        risk_level=risk,
+        reversible=False
+    )
+
+
+def create_git_action(operation: str, details: Dict[str, Any] = None) -> ActionRequest:
+    """Create an action request for git operation."""
+    return ActionRequest(
+        action_type=ActionType.GIT_OPERATION,
+        description=f"Git {operation}",
+        details=details or {},
+        risk_level="low" if operation in ["status", "log", "diff"] else "medium",
+        reversible=operation not in ["push", "force-push"]
+    )
+
+
+def create_tool_call_action(tool_name: str, args: Dict[str, Any] = None) -> ActionRequest:
+    """Create an action request for tool call."""
+    return ActionRequest(
+        action_type=ActionType.TOOL_CALL,
+        description=f"Call tool: {tool_name}",
+        details={
+            "tool": tool_name,
+            "args": args or {}
+        },
+        risk_level="low",
+        reversible=True
+    )
@@ -0,0 +1,468 @@
+"""
+CLI Compare Feature for PraisonAI.
+
+Compares different CLI command modes without performance impact.
+Only loaded when --compare flag is used.
+"""
+
+import time
+import json
+import logging
+from dataclasses import dataclass, field
+from typing import Dict, Any, List, Optional
+from datetime import datetime
+
+logger = logging.getLogger(__name__)
+
+COMPARE_MODES = {
+    "basic": {},
+    "tools": {"tools": "internet_search"},
+    "research": {"research": True},
+    "planning": {"planning": True},
+    "memory": {"memory": True},
+    "router": {"router": True},
+    "web_search": {"web_search": True},
+    "web_fetch": {"web_fetch": True},
+    "query_rewrite": {"query_rewrite": True},
+    "expand_prompt": {"expand_prompt": True},
+}
+
+
+def get_mode_config(mode_name: str, custom_args: Optional[Dict[str, Any]] = None) -> Dict[str, Any]:
+    """
+    Get configuration for a comparison mode.
+    
+    Args:
+        mode_name: Name of the mode
+        custom_args: Optional custom arguments to merge
+        
+    Returns:
+        Configuration dictionary for the mode
+    """
+    config = COMPARE_MODES.get(mode_name, {}).copy()
+    if custom_args:
+        config.update(custom_args)
+    return config
+
+
+def list_available_modes() -> List[str]:
+    """
+    List all available comparison modes.
+    
+    Returns:
+        List of mode names
+    """
+    return list(COMPARE_MODES.keys())
+
+
+def parse_modes(modes_str: str) -> List[str]:
+    """
+    Parse comma-separated modes string.
+    
+    Args:
+        modes_str: Comma-separated mode names
+        
+    Returns:
+        List of mode names
+    """
+    if not modes_str or not modes_str.strip():
+        return ["basic"]
+    
+    modes = [m.strip() for m in modes_str.split(",")]
+    return [m for m in modes if m]
+
+
+@dataclass
+class ModeResult:
+    """Result from running a single mode."""
+    mode: str
+    output: str
+    execution_time_ms: float
+    model_used: str
+    tokens: Optional[Dict[str, int]] = None
+    cost: Optional[float] = None
+    tools_used: List[str] = field(default_factory=list)
+    error: Optional[str] = None
+    timestamp: str = field(default_factory=lambda: datetime.now().isoformat())
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary."""
+        return {
+            "mode": self.mode,
+            "output_preview": self.output[:200] + "..." if len(self.output) > 200 else self.output,
+            "execution_time_ms": self.execution_time_ms,
+            "model_used": self.model_used,
+            "tokens": self.tokens,
+            "cost": self.cost,
+            "tools_used": self.tools_used,
+            "error": self.error,
+            "timestamp": self.timestamp,
+        }
+
+
+@dataclass
+class CompareResult:
+    """Result from comparing multiple modes."""
+    query: str
+    comparisons: List[ModeResult] = field(default_factory=list)
+    timestamp: str = field(default_factory=lambda: datetime.now().isoformat())
+    
+    def get_summary(self) -> Dict[str, Any]:
+        """Generate summary statistics."""
+        if not self.comparisons:
+            return {}
+        
+        times = [(c.mode, c.execution_time_ms) for c in self.comparisons if c.error is None]
+        if not times:
+            return {"error": "All modes failed"}
+        
+        sorted_by_time = sorted(times, key=lambda x: x[1])
+        
+        summary = {
+            "fastest": sorted_by_time[0][0],
+            "slowest": sorted_by_time[-1][0],
+            "fastest_time_ms": sorted_by_time[0][1],
+            "slowest_time_ms": sorted_by_time[-1][1],
+        }
+        
+        total_tokens = 0
+        total_cost = 0.0
+        for c in self.comparisons:
+            if c.tokens:
+                total_tokens += c.tokens.get("input", 0) + c.tokens.get("output", 0)
+            if c.cost:
+                total_cost += c.cost
+        
+        if total_tokens > 0:
+            summary["total_tokens"] = total_tokens
+        if total_cost > 0:
+            summary["total_cost"] = total_cost
+        
+        return summary
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary."""
+        return {
+            "query": self.query,
+            "timestamp": self.timestamp,
+            "comparisons": [c.to_dict() for c in self.comparisons],
+            "summary": self.get_summary(),
+        }
+    
+    def to_json(self) -> str:
+        """Convert to JSON string."""
+        return json.dumps(self.to_dict(), indent=2)
+
+
+def save_compare_result(result: CompareResult, path: str) -> bool:
+    """
+    Save comparison result to file.
+    
+    Args:
+        result: CompareResult to save
+        path: File path
+        
+    Returns:
+        True if successful
+    """
+    try:
+        from pathlib import Path
+        
+        file_path = Path(path)
+        file_path.parent.mkdir(parents=True, exist_ok=True)
+        file_path.write_text(result.to_json())
+        return True
+    except Exception as e:
+        logger.error(f"Failed to save comparison result: {e}")
+        return False
+
+
+def format_comparison_table(result: CompareResult) -> str:
+    """
+    Format comparison result as a table string.
+    
+    Args:
+        result: CompareResult to format
+        
+    Returns:
+        Formatted table string
+    """
+    try:
+        from rich.console import Console
+        from rich.table import Table
+        from io import StringIO
+        
+        console = Console(file=StringIO(), force_terminal=True)
+        
+        table = Table(title=f"Comparison: {result.query[:50]}...")
+        table.add_column("Mode", style="cyan")
+        table.add_column("Time (ms)", style="green")
+        table.add_column("Model", style="yellow")
+        table.add_column("Tools", style="magenta")
+        table.add_column("Status", style="bold")
+        
+        for c in result.comparisons:
+            status = "✅" if c.error is None else f"❌ {c.error[:20]}"
+            tools = ", ".join(c.tools_used) if c.tools_used else "-"
+            table.add_row(
+                c.mode,
+                f"{c.execution_time_ms:.1f}",
+                c.model_used,
+                tools,
+                status
+            )
+        
+        summary = result.get_summary()
+        if summary.get("fastest"):
+            table.add_section()
+            table.add_row(
+                "Summary",
+                f"Fastest: {summary['fastest']}",
+                "",
+                "",
+                f"Δ {summary['slowest_time_ms'] - summary['fastest_time_ms']:.1f}ms"
+            )
+        
+        console.print(table)
+        return console.file.getvalue()
+    except ImportError:
+        lines = [f"Comparison: {result.query}"]
+        lines.append("-" * 60)
+        for c in result.comparisons:
+            status = "OK" if c.error is None else f"ERROR: {c.error}"
+            lines.append(f"{c.mode}: {c.execution_time_ms:.1f}ms - {status}")
+        return "\n".join(lines)
+
+
+class CompareHandler:
+    """
+    Handler for CLI compare feature.
+    
+    Compares different CLI command modes and generates comparison reports.
+    """
+    
+    def __init__(self, verbose: bool = False):
+        """
+        Initialize the compare handler.
+        
+        Args:
+            verbose: Enable verbose output
+        """
+        self.verbose = verbose
+        self.results: List[ModeResult] = []
+    
+    def compare(
+        self,
+        query: str,
+        modes: List[str],
+        model: Optional[str] = None,
+        **kwargs
+    ) -> CompareResult:
+        """
+        Compare query execution across different modes.
+        
+        Args:
+            query: The query to execute
+            modes: List of mode names to compare
+            model: Optional model override
+            **kwargs: Additional arguments
+            
+        Returns:
+            CompareResult with all comparisons
+        """
+        result = CompareResult(query=query)
+        
+        for mode in modes:
+            if self.verbose:
+                logger.info(f"Running mode: {mode}")
+            
+            try:
+                mode_result = self._run_mode(query, mode, model=model, **kwargs)
+                result.comparisons.append(mode_result)
+            except Exception as e:
+                logger.error(f"Error running mode {mode}: {e}")
+                result.comparisons.append(ModeResult(
+                    mode=mode,
+                    output="",
+                    execution_time_ms=0,
+                    model_used=model or "unknown",
+                    error=str(e)
+                ))
+        
+        return result
+    
+    def _run_mode(
+        self,
+        query: str,
+        mode: str,
+        model: Optional[str] = None,
+        **kwargs
+    ) -> ModeResult:
+        """
+        Run a single mode and capture results.
+        
+        Args:
+            query: The query to execute
+            mode: Mode name
+            model: Optional model override
+            **kwargs: Additional arguments
+            
+        Returns:
+            ModeResult with execution details
+        """
+        config = get_mode_config(mode, kwargs)
+        
+        if model:
+            config["llm"] = model
+        
+        start_time = time.perf_counter()
+        output = ""
+        tokens = None
+        cost = None
+        tools_used = []
+        error = None
+        model_used = model or "gpt-4o-mini"
+        
+        try:
+            from praisonaiagents import Agent
+            
+            agent_config = {
+                "name": f"CompareAgent_{mode}",
+                "role": "Assistant",
+                "goal": "Complete the given task",
+                "backstory": "You are a helpful AI assistant",
+                "verbose": False,
+            }
+            
+            if config.get("llm"):
+                agent_config["llm"] = config["llm"]
+                model_used = config["llm"]
+            
+            if config.get("tools"):
+                tools_list = self._load_tools(config["tools"])
+                if tools_list:
+                    agent_config["tools"] = tools_list
+            
+            if config.get("planning"):
+                agent_config["planning"] = True
+            
+            if config.get("memory"):
+                agent_config["memory"] = True
+            
+            if config.get("web_search"):
+                agent_config["web_search"] = True
+            
+            if config.get("web_fetch"):
+                agent_config["web_fetch"] = True
+            
+            agent = Agent(**agent_config)
+            result = agent.start(query)
+            
+            if hasattr(result, 'raw'):
+                output = str(result.raw)
+            else:
+                output = str(result)
+            
+            if hasattr(agent, 'tools') and agent.tools:
+                for tool in agent.tools:
+                    if hasattr(tool, '__name__'):
+                        tools_used.append(tool.__name__)
+                    elif hasattr(tool, 'name'):
+                        tools_used.append(tool.name)
+            
+        except Exception as e:
+            error = str(e)
+            logger.error(f"Mode {mode} failed: {e}")
+        
+        end_time = time.perf_counter()
+        execution_time_ms = (end_time - start_time) * 1000
+        
+        return ModeResult(
+            mode=mode,
+            output=output,
+            execution_time_ms=execution_time_ms,
+            model_used=model_used,
+            tokens=tokens,
+            cost=cost,
+            tools_used=tools_used,
+            error=error
+        )
+    
+    def _load_tools(self, tools_str: str) -> List:
+        """
+        Load tools from string specification.
+        
+        Args:
+            tools_str: Comma-separated tool names
+            
+        Returns:
+            List of tool functions
+        """
+        tools = []
+        tool_names = [t.strip() for t in tools_str.split(",")]
+        
+        for name in tool_names:
+            try:
+                if name == "internet_search":
+                    from praisonaiagents.tools import internet_search
+                    tools.append(internet_search)
+                elif name == "calculator":
+                    def calculator(expression: str) -> str:
+                        """Evaluate a mathematical expression."""
+                        try:
+                            return str(eval(expression))
+                        except Exception as e:
+                            return f"Error: {e}"
+                    tools.append(calculator)
+            except ImportError:
+                logger.warning(f"Could not load tool: {name}")
+        
+        return tools
+    
+    def print_result(self, result: CompareResult) -> None:
+        """
+        Print comparison result to console.
+        
+        Args:
+            result: CompareResult to print
+        """
+        table_str = format_comparison_table(result)
+        print(table_str)
+    
+    def execute(
+        self,
+        query: str,
+        modes_str: str,
+        model: Optional[str] = None,
+        output_path: Optional[str] = None,
+        **kwargs
+    ) -> CompareResult:
+        """
+        Execute comparison from CLI arguments.
+        
+        Args:
+            query: The query to execute
+            modes_str: Comma-separated mode names
+            model: Optional model override
+            output_path: Optional path to save results
+            **kwargs: Additional arguments
+            
+        Returns:
+            CompareResult with all comparisons
+        """
+        modes = parse_modes(modes_str)
+        
+        if self.verbose:
+            print(f"[bold cyan]Comparing modes: {', '.join(modes)}[/bold cyan]")
+        
+        result = self.compare(query, modes, model=model, **kwargs)
+        
+        self.print_result(result)
+        
+        if output_path:
+            if save_compare_result(result, output_path):
+                print(f"[green]Results saved to: {output_path}[/green]")
+            else:
+                print(f"[red]Failed to save results to: {output_path}[/red]")
+        
+        return result
@@ -0,0 +1,514 @@
+"""
+Cost Tracking System for PraisonAI CLI.
+
+Inspired by Aider's cost tracking and Gemini CLI's stats command.
+Provides real-time cost and token usage tracking.
+
+Architecture:
+- CostTracker: Tracks tokens and costs per session
+- ModelPricing: Pricing data for different models
+- UsageStats: Aggregated usage statistics
+"""
+
+from dataclasses import dataclass, field
+from typing import Any, Dict, List, Optional
+from datetime import datetime
+import logging
+import json
+
+logger = logging.getLogger(__name__)
+
+
+# ============================================================================
+# Model Pricing Data
+# ============================================================================
+
+@dataclass
+class ModelPricing:
+    """
+    Pricing information for a model.
+    
+    Prices are per 1M tokens (as commonly quoted).
+    """
+    model_name: str
+    input_price_per_1m: float  # Price per 1M input tokens
+    output_price_per_1m: float  # Price per 1M output tokens
+    context_window: int = 128000
+    
+    def calculate_cost(self, input_tokens: int, output_tokens: int) -> float:
+        """Calculate cost for given token counts."""
+        input_cost = (input_tokens / 1_000_000) * self.input_price_per_1m
+        output_cost = (output_tokens / 1_000_000) * self.output_price_per_1m
+        return input_cost + output_cost
+
+
+# Default pricing for common models (as of late 2024)
+DEFAULT_PRICING: Dict[str, ModelPricing] = {
+    # OpenAI models
+    "gpt-4o": ModelPricing("gpt-4o", 2.50, 10.00, 128000),
+    "gpt-4o-mini": ModelPricing("gpt-4o-mini", 0.15, 0.60, 128000),
+    "gpt-4-turbo": ModelPricing("gpt-4-turbo", 10.00, 30.00, 128000),
+    "gpt-4": ModelPricing("gpt-4", 30.00, 60.00, 8192),
+    "gpt-3.5-turbo": ModelPricing("gpt-3.5-turbo", 0.50, 1.50, 16385),
+    "o1": ModelPricing("o1", 15.00, 60.00, 200000),
+    "o1-mini": ModelPricing("o1-mini", 3.00, 12.00, 128000),
+    "o1-preview": ModelPricing("o1-preview", 15.00, 60.00, 128000),
+    "o3-mini": ModelPricing("o3-mini", 1.10, 4.40, 200000),
+    
+    # Anthropic models
+    "claude-3-5-sonnet-20241022": ModelPricing("claude-3-5-sonnet", 3.00, 15.00, 200000),
+    "claude-3-5-sonnet": ModelPricing("claude-3-5-sonnet", 3.00, 15.00, 200000),
+    "claude-3-opus": ModelPricing("claude-3-opus", 15.00, 75.00, 200000),
+    "claude-3-sonnet": ModelPricing("claude-3-sonnet", 3.00, 15.00, 200000),
+    "claude-3-haiku": ModelPricing("claude-3-haiku", 0.25, 1.25, 200000),
+    
+    # Google models
+    "gemini-2.0-flash": ModelPricing("gemini-2.0-flash", 0.10, 0.40, 1000000),
+    "gemini-1.5-pro": ModelPricing("gemini-1.5-pro", 1.25, 5.00, 2000000),
+    "gemini-1.5-flash": ModelPricing("gemini-1.5-flash", 0.075, 0.30, 1000000),
+    
+    # Default fallback
+    "default": ModelPricing("default", 1.00, 3.00, 128000),
+}
+
+
+def get_pricing(model_name: str) -> ModelPricing:
+    """Get pricing for a model, with fallback to default."""
+    # Try exact match
+    if model_name in DEFAULT_PRICING:
+        return DEFAULT_PRICING[model_name]
+    
+    # Try partial match
+    model_lower = model_name.lower()
+    for key, pricing in DEFAULT_PRICING.items():
+        if key in model_lower or model_lower in key:
+            return pricing
+    
+    # Return default
+    logger.debug(f"No pricing found for model '{model_name}', using default")
+    return DEFAULT_PRICING["default"]
+
+
+# ============================================================================
+# Usage Statistics
+# ============================================================================
+
+@dataclass
+class TokenUsage:
+    """Token usage for a single request."""
+    input_tokens: int = 0
+    output_tokens: int = 0
+    total_tokens: int = 0
+    cached_tokens: int = 0
+    
+    def __post_init__(self):
+        if self.total_tokens == 0:
+            self.total_tokens = self.input_tokens + self.output_tokens
+
+
+@dataclass
+class RequestStats:
+    """Statistics for a single request."""
+    timestamp: datetime
+    model: str
+    usage: TokenUsage
+    cost: float
+    duration_ms: float = 0.0
+    prompt_preview: str = ""
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary for serialization."""
+        return {
+            "timestamp": self.timestamp.isoformat(),
+            "model": self.model,
+            "input_tokens": self.usage.input_tokens,
+            "output_tokens": self.usage.output_tokens,
+            "total_tokens": self.usage.total_tokens,
+            "cached_tokens": self.usage.cached_tokens,
+            "cost": self.cost,
+            "duration_ms": self.duration_ms,
+        }
+
+
+@dataclass
+class SessionStats:
+    """Aggregated statistics for a session."""
+    session_id: str
+    start_time: datetime
+    end_time: Optional[datetime] = None
+    
+    total_requests: int = 0
+    total_input_tokens: int = 0
+    total_output_tokens: int = 0
+    total_tokens: int = 0
+    total_cached_tokens: int = 0
+    total_cost: float = 0.0
+    total_duration_ms: float = 0.0
+    
+    models_used: Dict[str, int] = field(default_factory=dict)
+    
+    def add_request(self, stats: RequestStats) -> None:
+        """Add request statistics to session totals."""
+        self.total_requests += 1
+        self.total_input_tokens += stats.usage.input_tokens
+        self.total_output_tokens += stats.usage.output_tokens
+        self.total_tokens += stats.usage.total_tokens
+        self.total_cached_tokens += stats.usage.cached_tokens
+        self.total_cost += stats.cost
+        self.total_duration_ms += stats.duration_ms
+        
+        # Track model usage
+        if stats.model not in self.models_used:
+            self.models_used[stats.model] = 0
+        self.models_used[stats.model] += 1
+    
+    @property
+    def duration_seconds(self) -> float:
+        """Get session duration in seconds."""
+        end = self.end_time or datetime.now()
+        return (end - self.start_time).total_seconds()
+    
+    @property
+    def avg_tokens_per_request(self) -> float:
+        """Get average tokens per request."""
+        if self.total_requests == 0:
+            return 0.0
+        return self.total_tokens / self.total_requests
+    
+    @property
+    def avg_cost_per_request(self) -> float:
+        """Get average cost per request."""
+        if self.total_requests == 0:
+            return 0.0
+        return self.total_cost / self.total_requests
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary for serialization."""
+        return {
+            "session_id": self.session_id,
+            "start_time": self.start_time.isoformat(),
+            "end_time": self.end_time.isoformat() if self.end_time else None,
+            "duration_seconds": self.duration_seconds,
+            "total_requests": self.total_requests,
+            "total_input_tokens": self.total_input_tokens,
+            "total_output_tokens": self.total_output_tokens,
+            "total_tokens": self.total_tokens,
+            "total_cached_tokens": self.total_cached_tokens,
+            "total_cost": self.total_cost,
+            "avg_tokens_per_request": self.avg_tokens_per_request,
+            "avg_cost_per_request": self.avg_cost_per_request,
+            "models_used": self.models_used,
+        }
+
+
+# ============================================================================
+# Cost Tracker
+# ============================================================================
+
+class CostTracker:
+    """
+    Tracks costs and token usage for a session.
+    
+    Features:
+    - Real-time cost tracking
+    - Per-model statistics
+    - Session history
+    - Export to JSON
+    """
+    
+    def __init__(
+        self,
+        session_id: Optional[str] = None,
+        pricing: Optional[Dict[str, ModelPricing]] = None,
+        verbose: bool = False
+    ):
+        self.session_id = session_id or self._generate_session_id()
+        self.pricing = pricing or DEFAULT_PRICING
+        self.verbose = verbose
+        
+        self.session_stats = SessionStats(
+            session_id=self.session_id,
+            start_time=datetime.now()
+        )
+        self.request_history: List[RequestStats] = []
+    
+    def _generate_session_id(self) -> str:
+        """Generate a unique session ID."""
+        import uuid
+        return str(uuid.uuid4())[:8]
+    
+    def track_request(
+        self,
+        model: str,
+        input_tokens: int,
+        output_tokens: int,
+        cached_tokens: int = 0,
+        duration_ms: float = 0.0,
+        prompt_preview: str = ""
+    ) -> RequestStats:
+        """
+        Track a single request.
+        
+        Returns:
+            RequestStats for the tracked request
+        """
+        usage = TokenUsage(
+            input_tokens=input_tokens,
+            output_tokens=output_tokens,
+            cached_tokens=cached_tokens
+        )
+        
+        # Calculate cost
+        pricing = get_pricing(model)
+        cost = pricing.calculate_cost(input_tokens, output_tokens)
+        
+        # Create request stats
+        stats = RequestStats(
+            timestamp=datetime.now(),
+            model=model,
+            usage=usage,
+            cost=cost,
+            duration_ms=duration_ms,
+            prompt_preview=prompt_preview[:100] if prompt_preview else ""
+        )
+        
+        # Update session stats
+        self.session_stats.add_request(stats)
+        self.request_history.append(stats)
+        
+        if self.verbose:
+            logger.info(
+                f"Request tracked: {model} - "
+                f"{usage.total_tokens} tokens, ${cost:.4f}"
+            )
+        
+        return stats
+    
+    def track_from_response(
+        self,
+        model: str,
+        response: Any,
+        duration_ms: float = 0.0
+    ) -> Optional[RequestStats]:
+        """
+        Track request from LLM response object.
+        
+        Handles various response formats from different providers.
+        """
+        try:
+            # Try to extract usage from response
+            usage = None
+            
+            # OpenAI format
+            if hasattr(response, 'usage'):
+                usage = response.usage
+                input_tokens = getattr(usage, 'prompt_tokens', 0)
+                output_tokens = getattr(usage, 'completion_tokens', 0)
+                cached_tokens = getattr(usage, 'cached_tokens', 0)
+            # Dict format
+            elif isinstance(response, dict) and 'usage' in response:
+                usage = response['usage']
+                input_tokens = usage.get('prompt_tokens', 0)
+                output_tokens = usage.get('completion_tokens', 0)
+                cached_tokens = usage.get('cached_tokens', 0)
+            else:
+                logger.debug("Could not extract usage from response")
+                return None
+            
+            return self.track_request(
+                model=model,
+                input_tokens=input_tokens,
+                output_tokens=output_tokens,
+                cached_tokens=cached_tokens,
+                duration_ms=duration_ms
+            )
+        except Exception as e:
+            logger.error(f"Error tracking response: {e}")
+            return None
+    
+    def get_current_stats(self) -> SessionStats:
+        """Get current session statistics."""
+        return self.session_stats
+    
+    def get_total_cost(self) -> float:
+        """Get total cost for the session."""
+        return self.session_stats.total_cost
+    
+    def get_total_tokens(self) -> int:
+        """Get total tokens for the session."""
+        return self.session_stats.total_tokens
+    
+    def get_request_count(self) -> int:
+        """Get total request count."""
+        return self.session_stats.total_requests
+    
+    def format_summary(self) -> str:
+        """Format a summary of the session statistics."""
+        stats = self.session_stats
+        
+        lines = [
+            f"Session: {stats.session_id}",
+            f"Duration: {stats.duration_seconds:.1f}s",
+            f"Requests: {stats.total_requests}",
+            "",
+            "Tokens:",
+            f"  Input:  {stats.total_input_tokens:,}",
+            f"  Output: {stats.total_output_tokens:,}",
+            f"  Total:  {stats.total_tokens:,}",
+            f"  Cached: {stats.total_cached_tokens:,}",
+            "",
+            f"Cost: ${stats.total_cost:.4f}",
+            f"Avg per request: ${stats.avg_cost_per_request:.4f}",
+        ]
+        
+        if stats.models_used:
+            lines.append("")
+            lines.append("Models used:")
+            for model, count in stats.models_used.items():
+                lines.append(f"  {model}: {count} requests")
+        
+        return "\n".join(lines)
+    
+    def print_summary(self) -> None:
+        """Print a formatted summary to console."""
+        from rich.console import Console
+        from rich.panel import Panel
+        from rich.table import Table
+        
+        console = Console()
+        stats = self.session_stats
+        
+        # Create summary table
+        table = Table(show_header=True, header_style="bold cyan")
+        table.add_column("Metric")
+        table.add_column("Value", justify="right")
+        
+        table.add_row("Session ID", stats.session_id)
+        table.add_row("Duration", f"{stats.duration_seconds:.1f}s")
+        table.add_row("Requests", str(stats.total_requests))
+        table.add_row("", "")
+        table.add_row("Input Tokens", f"{stats.total_input_tokens:,}")
+        table.add_row("Output Tokens", f"{stats.total_output_tokens:,}")
+        table.add_row("Total Tokens", f"{stats.total_tokens:,}")
+        table.add_row("Cached Tokens", f"{stats.total_cached_tokens:,}")
+        table.add_row("", "")
+        table.add_row("Total Cost", f"${stats.total_cost:.4f}")
+        table.add_row("Avg Cost/Request", f"${stats.avg_cost_per_request:.4f}")
+        
+        console.print(Panel(table, title="💰 Session Statistics", border_style="green"))
+        
+        # Print model breakdown if multiple models used
+        if len(stats.models_used) > 1:
+            model_table = Table(show_header=True, header_style="bold cyan")
+            model_table.add_column("Model")
+            model_table.add_column("Requests", justify="right")
+            
+            for model, count in stats.models_used.items():
+                model_table.add_row(model, str(count))
+            
+            console.print(Panel(model_table, title="Models Used", border_style="blue"))
+    
+    def export_json(self, filepath: Optional[str] = None) -> str:
+        """
+        Export session data to JSON.
+        
+        Args:
+            filepath: Optional file path to write to
+            
+        Returns:
+            JSON string
+        """
+        data = {
+            "session": self.session_stats.to_dict(),
+            "requests": [r.to_dict() for r in self.request_history]
+        }
+        
+        json_str = json.dumps(data, indent=2)
+        
+        if filepath:
+            with open(filepath, 'w') as f:
+                f.write(json_str)
+            logger.info(f"Exported session data to {filepath}")
+        
+        return json_str
+    
+    def end_session(self) -> SessionStats:
+        """Mark the session as ended."""
+        self.session_stats.end_time = datetime.now()
+        return self.session_stats
+
+
+# ============================================================================
+# CLI Integration Handler
+# ============================================================================
+
+class CostTrackerHandler:
+    """
+    Handler for integrating cost tracking with PraisonAI CLI.
+    """
+    
+    def __init__(self, verbose: bool = False):
+        self.verbose = verbose
+        self._tracker: Optional[CostTracker] = None
+    
+    @property
+    def feature_name(self) -> str:
+        return "cost_tracker"
+    
+    def initialize(self, session_id: Optional[str] = None) -> CostTracker:
+        """Initialize the cost tracker."""
+        self._tracker = CostTracker(
+            session_id=session_id,
+            verbose=self.verbose
+        )
+        
+        if self.verbose:
+            from rich import print as rprint
+            rprint(f"[cyan]Cost tracking enabled for session: {self._tracker.session_id}[/cyan]")
+        
+        return self._tracker
+    
+    def get_tracker(self) -> Optional[CostTracker]:
+        """Get the current cost tracker."""
+        return self._tracker
+    
+    def track_request(
+        self,
+        model: str,
+        input_tokens: int,
+        output_tokens: int,
+        **kwargs
+    ) -> Optional[RequestStats]:
+        """Track a request."""
+        if not self._tracker:
+            self._tracker = self.initialize()
+        
+        return self._tracker.track_request(
+            model=model,
+            input_tokens=input_tokens,
+            output_tokens=output_tokens,
+            **kwargs
+        )
+    
+    def get_summary(self) -> Dict[str, Any]:
+        """Get session summary as dict."""
+        if not self._tracker:
+            return {}
+        return self._tracker.session_stats.to_dict()
+    
+    def print_summary(self) -> None:
+        """Print session summary."""
+        if self._tracker:
+            self._tracker.print_summary()
+    
+    def get_cost(self) -> float:
+        """Get total cost."""
+        if self._tracker:
+            return self._tracker.get_total_cost()
+        return 0.0
+    
+    def get_tokens(self) -> int:
+        """Get total tokens."""
+        if self._tracker:
+            return self._tracker.get_total_tokens()
+        return 0
@@ -0,0 +1,471 @@
+"""
+Evaluation CLI feature for PraisonAI.
+
+Provides CLI commands for running agent evaluations.
+"""
+
+import os
+import json
+import logging
+from typing import Optional, List, Dict, Any
+
+logger = logging.getLogger(__name__)
+
+
+class EvalHandler:
+    """Handler for evaluation CLI commands."""
+    
+    def __init__(self, verbose: bool = False):
+        """
+        Initialize the evaluation handler.
+        
+        Args:
+            verbose: Enable verbose output
+        """
+        self.verbose = verbose
+    
+    def run_accuracy(
+        self,
+        agent_file: str,
+        input_text: str,
+        expected_output: str,
+        iterations: int = 1,
+        model: Optional[str] = None,
+        output_file: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """
+        Run accuracy evaluation on an agent.
+        
+        Args:
+            agent_file: Path to agents.yaml file
+            input_text: Input to provide to the agent
+            expected_output: Expected output to compare against
+            iterations: Number of evaluation iterations
+            model: LLM model for judging
+            output_file: Path to save results
+            
+        Returns:
+            Evaluation result dictionary
+        """
+        try:
+            from praisonaiagents.eval import AccuracyEvaluator
+            from praisonai.agents_generator import AgentsGenerator
+        except ImportError as e:
+            logger.error(f"Failed to import evaluation modules: {e}")
+            return {"error": str(e)}
+        
+        try:
+            generator = AgentsGenerator(agent_file)
+            agents = generator.generate_agents()
+            
+            if not agents:
+                return {"error": "No agents found in configuration"}
+            
+            agent = agents[0] if isinstance(agents, list) else agents
+            
+            evaluator = AccuracyEvaluator(
+                agent=agent,
+                input_text=input_text,
+                expected_output=expected_output,
+                num_iterations=iterations,
+                model=model,
+                save_results_path=output_file,
+                verbose=self.verbose
+            )
+            
+            result = evaluator.run(print_summary=True)
+            return result.to_dict()
+            
+        except Exception as e:
+            logger.error(f"Accuracy evaluation failed: {e}")
+            return {"error": str(e)}
+    
+    def run_performance(
+        self,
+        agent_file: str,
+        input_text: str = "Hello",
+        iterations: int = 10,
+        warmup: int = 2,
+        track_memory: bool = True,
+        output_file: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """
+        Run performance evaluation on an agent.
+        
+        Args:
+            agent_file: Path to agents.yaml file
+            input_text: Input to provide to the agent
+            iterations: Number of benchmark iterations
+            warmup: Number of warmup runs
+            track_memory: Whether to track memory usage
+            output_file: Path to save results
+            
+        Returns:
+            Evaluation result dictionary
+        """
+        try:
+            from praisonaiagents.eval import PerformanceEvaluator
+            from praisonai.agents_generator import AgentsGenerator
+        except ImportError as e:
+            logger.error(f"Failed to import evaluation modules: {e}")
+            return {"error": str(e)}
+        
+        try:
+            generator = AgentsGenerator(agent_file)
+            agents = generator.generate_agents()
+            
+            if not agents:
+                return {"error": "No agents found in configuration"}
+            
+            agent = agents[0] if isinstance(agents, list) else agents
+            
+            evaluator = PerformanceEvaluator(
+                agent=agent,
+                input_text=input_text,
+                num_iterations=iterations,
+                warmup_runs=warmup,
+                track_memory=track_memory,
+                save_results_path=output_file,
+                verbose=self.verbose
+            )
+            
+            result = evaluator.run(print_summary=True)
+            return result.to_dict()
+            
+        except Exception as e:
+            logger.error(f"Performance evaluation failed: {e}")
+            return {"error": str(e)}
+    
+    def run_reliability(
+        self,
+        agent_file: str,
+        input_text: str,
+        expected_tools: List[str],
+        forbidden_tools: Optional[List[str]] = None,
+        output_file: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """
+        Run reliability evaluation on an agent.
+        
+        Args:
+            agent_file: Path to agents.yaml file
+            input_text: Input to provide to the agent
+            expected_tools: List of tools that should be called
+            forbidden_tools: List of tools that should NOT be called
+            output_file: Path to save results
+            
+        Returns:
+            Evaluation result dictionary
+        """
+        try:
+            from praisonaiagents.eval import ReliabilityEvaluator
+            from praisonai.agents_generator import AgentsGenerator
+        except ImportError as e:
+            logger.error(f"Failed to import evaluation modules: {e}")
+            return {"error": str(e)}
+        
+        try:
+            generator = AgentsGenerator(agent_file)
+            agents = generator.generate_agents()
+            
+            if not agents:
+                return {"error": "No agents found in configuration"}
+            
+            agent = agents[0] if isinstance(agents, list) else agents
+            
+            evaluator = ReliabilityEvaluator(
+                agent=agent,
+                input_text=input_text,
+                expected_tools=expected_tools,
+                forbidden_tools=forbidden_tools,
+                save_results_path=output_file,
+                verbose=self.verbose
+            )
+            
+            result = evaluator.run(print_summary=True)
+            return result.to_dict()
+            
+        except Exception as e:
+            logger.error(f"Reliability evaluation failed: {e}")
+            return {"error": str(e)}
+    
+    def run_criteria(
+        self,
+        agent_file: str,
+        input_text: str,
+        criteria: str,
+        scoring_type: str = "numeric",
+        threshold: float = 7.0,
+        iterations: int = 1,
+        model: Optional[str] = None,
+        output_file: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """
+        Run criteria-based evaluation on an agent.
+        
+        Args:
+            agent_file: Path to agents.yaml file
+            input_text: Input to provide to the agent
+            criteria: Criteria to evaluate against
+            scoring_type: "numeric" or "binary"
+            threshold: Score threshold for passing (numeric mode)
+            iterations: Number of evaluation iterations
+            model: LLM model for judging
+            output_file: Path to save results
+            
+        Returns:
+            Evaluation result dictionary
+        """
+        try:
+            from praisonaiagents.eval import CriteriaEvaluator
+            from praisonai.agents_generator import AgentsGenerator
+        except ImportError as e:
+            logger.error(f"Failed to import evaluation modules: {e}")
+            return {"error": str(e)}
+        
+        try:
+            generator = AgentsGenerator(agent_file)
+            agents = generator.generate_agents()
+            
+            if not agents:
+                return {"error": "No agents found in configuration"}
+            
+            agent = agents[0] if isinstance(agents, list) else agents
+            
+            evaluator = CriteriaEvaluator(
+                criteria=criteria,
+                agent=agent,
+                input_text=input_text,
+                scoring_type=scoring_type,
+                threshold=threshold,
+                num_iterations=iterations,
+                model=model,
+                save_results_path=output_file,
+                verbose=self.verbose
+            )
+            
+            result = evaluator.run(print_summary=True)
+            return result.to_dict()
+            
+        except Exception as e:
+            logger.error(f"Criteria evaluation failed: {e}")
+            return {"error": str(e)}
+    
+    def run_batch(
+        self,
+        agent_file: str,
+        test_file: str,
+        eval_type: str = "accuracy",
+        output_file: Optional[str] = None
+    ) -> Dict[str, Any]:
+        """
+        Run batch evaluation from a test file.
+        
+        Args:
+            agent_file: Path to agents.yaml file
+            test_file: Path to JSON test file with test cases
+            eval_type: Type of evaluation ("accuracy", "criteria")
+            output_file: Path to save results
+            
+        Returns:
+            Batch evaluation results
+        """
+        try:
+            with open(test_file, 'r') as f:
+                test_cases = json.load(f)
+        except Exception as e:
+            return {"error": f"Failed to load test file: {e}"}
+        
+        results = []
+        for i, test_case in enumerate(test_cases):
+            if self.verbose:
+                print(f"Running test case {i + 1}/{len(test_cases)}")
+            
+            if eval_type == "accuracy":
+                result = self.run_accuracy(
+                    agent_file=agent_file,
+                    input_text=test_case.get("input", ""),
+                    expected_output=test_case.get("expected", ""),
+                    iterations=test_case.get("iterations", 1)
+                )
+            elif eval_type == "criteria":
+                result = self.run_criteria(
+                    agent_file=agent_file,
+                    input_text=test_case.get("input", ""),
+                    criteria=test_case.get("criteria", ""),
+                    scoring_type=test_case.get("scoring_type", "numeric"),
+                    threshold=test_case.get("threshold", 7.0)
+                )
+            else:
+                result = {"error": f"Unknown eval type: {eval_type}"}
+            
+            results.append({
+                "test_case": i + 1,
+                "input": test_case.get("input", ""),
+                "result": result
+            })
+        
+        batch_result = {
+            "total_tests": len(test_cases),
+            "eval_type": eval_type,
+            "results": results
+        }
+        
+        if output_file:
+            try:
+                with open(output_file, 'w') as f:
+                    json.dump(batch_result, f, indent=2)
+            except Exception as e:
+                logger.warning(f"Failed to save batch results: {e}")
+        
+        return batch_result
+
+
+def handle_eval_command(args) -> None:
+    """
+    Handle the eval CLI command.
+    
+    Args:
+        args: Parsed command line arguments
+    """
+    handler = EvalHandler(verbose=getattr(args, 'verbose', False))
+    
+    eval_type = getattr(args, 'eval_type', 'accuracy')
+    agent_file = getattr(args, 'agent', 'agents.yaml')
+    output_file = getattr(args, 'output', None)
+    
+    if eval_type == 'accuracy':
+        result = handler.run_accuracy(
+            agent_file=agent_file,
+            input_text=getattr(args, 'input', ''),
+            expected_output=getattr(args, 'expected', ''),
+            iterations=getattr(args, 'iterations', 1),
+            model=getattr(args, 'model', None),
+            output_file=output_file
+        )
+    elif eval_type == 'performance':
+        result = handler.run_performance(
+            agent_file=agent_file,
+            input_text=getattr(args, 'input', 'Hello'),
+            iterations=getattr(args, 'iterations', 10),
+            warmup=getattr(args, 'warmup', 2),
+            track_memory=getattr(args, 'memory', True),
+            output_file=output_file
+        )
+    elif eval_type == 'reliability':
+        expected_tools = getattr(args, 'expected_tools', '').split(',')
+        forbidden_tools = getattr(args, 'forbidden_tools', '')
+        forbidden_tools = forbidden_tools.split(',') if forbidden_tools else None
+        
+        result = handler.run_reliability(
+            agent_file=agent_file,
+            input_text=getattr(args, 'input', ''),
+            expected_tools=expected_tools,
+            forbidden_tools=forbidden_tools,
+            output_file=output_file
+        )
+    elif eval_type == 'criteria':
+        result = handler.run_criteria(
+            agent_file=agent_file,
+            input_text=getattr(args, 'input', ''),
+            criteria=getattr(args, 'criteria', ''),
+            scoring_type=getattr(args, 'scoring', 'numeric'),
+            threshold=getattr(args, 'threshold', 7.0),
+            iterations=getattr(args, 'iterations', 1),
+            model=getattr(args, 'model', None),
+            output_file=output_file
+        )
+    elif eval_type == 'batch':
+        result = handler.run_batch(
+            agent_file=agent_file,
+            test_file=getattr(args, 'test_file', ''),
+            eval_type=getattr(args, 'batch_type', 'accuracy'),
+            output_file=output_file
+        )
+    else:
+        print(f"Unknown evaluation type: {eval_type}")
+        return
+    
+    if 'error' in result:
+        print(f"Error: {result['error']}")
+    elif not getattr(args, 'quiet', False):
+        print(json.dumps(result, indent=2))
+
+
+def add_eval_parser(subparsers) -> None:
+    """
+    Add eval subcommand parser.
+    
+    Args:
+        subparsers: Argument parser subparsers
+    """
+    eval_parser = subparsers.add_parser(
+        'eval',
+        help='Run agent evaluations'
+    )
+    
+    eval_subparsers = eval_parser.add_subparsers(dest='eval_type')
+    
+    accuracy_parser = eval_subparsers.add_parser(
+        'accuracy',
+        help='Run accuracy evaluation'
+    )
+    accuracy_parser.add_argument('--agent', '-a', default='agents.yaml', help='Agent config file')
+    accuracy_parser.add_argument('--input', '-i', required=True, help='Input text')
+    accuracy_parser.add_argument('--expected', '-e', required=True, help='Expected output')
+    accuracy_parser.add_argument('--iterations', '-n', type=int, default=1, help='Number of iterations')
+    accuracy_parser.add_argument('--model', '-m', help='Judge model')
+    accuracy_parser.add_argument('--output', '-o', help='Output file')
+    accuracy_parser.add_argument('--verbose', '-v', action='store_true', help='Verbose output')
+    accuracy_parser.add_argument('--quiet', '-q', action='store_true', help='Suppress JSON output')
+    
+    perf_parser = eval_subparsers.add_parser(
+        'performance',
+        help='Run performance evaluation'
+    )
+    perf_parser.add_argument('--agent', '-a', default='agents.yaml', help='Agent config file')
+    perf_parser.add_argument('--input', '-i', default='Hello', help='Input text')
+    perf_parser.add_argument('--iterations', '-n', type=int, default=10, help='Number of iterations')
+    perf_parser.add_argument('--warmup', '-w', type=int, default=2, help='Warmup runs')
+    perf_parser.add_argument('--memory', action='store_true', default=True, help='Track memory')
+    perf_parser.add_argument('--output', '-o', help='Output file')
+    perf_parser.add_argument('--verbose', '-v', action='store_true', help='Verbose output')
+    perf_parser.add_argument('--quiet', '-q', action='store_true', help='Suppress JSON output')
+    
+    rel_parser = eval_subparsers.add_parser(
+        'reliability',
+        help='Run reliability evaluation'
+    )
+    rel_parser.add_argument('--agent', '-a', default='agents.yaml', help='Agent config file')
+    rel_parser.add_argument('--input', '-i', required=True, help='Input text')
+    rel_parser.add_argument('--expected-tools', '-t', required=True, help='Expected tools (comma-separated)')
+    rel_parser.add_argument('--forbidden-tools', '-f', help='Forbidden tools (comma-separated)')
+    rel_parser.add_argument('--output', '-o', help='Output file')
+    rel_parser.add_argument('--verbose', '-v', action='store_true', help='Verbose output')
+    rel_parser.add_argument('--quiet', '-q', action='store_true', help='Suppress JSON output')
+    
+    criteria_parser = eval_subparsers.add_parser(
+        'criteria',
+        help='Run criteria-based evaluation'
+    )
+    criteria_parser.add_argument('--agent', '-a', default='agents.yaml', help='Agent config file')
+    criteria_parser.add_argument('--input', '-i', required=True, help='Input text')
+    criteria_parser.add_argument('--criteria', '-c', required=True, help='Evaluation criteria')
+    criteria_parser.add_argument('--scoring', '-s', choices=['numeric', 'binary'], default='numeric', help='Scoring type')
+    criteria_parser.add_argument('--threshold', type=float, default=7.0, help='Pass threshold')
+    criteria_parser.add_argument('--iterations', '-n', type=int, default=1, help='Number of iterations')
+    criteria_parser.add_argument('--model', '-m', help='Judge model')
+    criteria_parser.add_argument('--output', '-o', help='Output file')
+    criteria_parser.add_argument('--verbose', '-v', action='store_true', help='Verbose output')
+    criteria_parser.add_argument('--quiet', '-q', action='store_true', help='Suppress JSON output')
+    
+    batch_parser = eval_subparsers.add_parser(
+        'batch',
+        help='Run batch evaluation from test file'
+    )
+    batch_parser.add_argument('--agent', '-a', default='agents.yaml', help='Agent config file')
+    batch_parser.add_argument('--test-file', '-t', required=True, help='JSON test file')
+    batch_parser.add_argument('--batch-type', '-b', choices=['accuracy', 'criteria'], default='accuracy', help='Evaluation type')
+    batch_parser.add_argument('--output', '-o', help='Output file')
+    batch_parser.add_argument('--verbose', '-v', action='store_true', help='Verbose output')
+    batch_parser.add_argument('--quiet', '-q', action='store_true', help='Suppress JSON output')
Author	SHA1	Message	Date
Mervin Praison	5a13571fdd	Release v2.3.75	2025-12-22 19:11:32 +00:00
Mervin Praison	e3b48c2526	Release v2.3.74	2025-12-22 17:01:50 +00:00
Mervin Praison	8d32238b4c	Release v2.3.73	2025-12-22 15:47:51 +00:00
Mervin Praison	c17b1298bb	Reorganize README feature table with improved categorization and hierarchy	2025-12-22 14:50:37 +00:00
Mervin Praison	ec6c150315	Release v2.3.72	2025-12-22 13:44:48 +00:00
Mervin Praison	3d26cedd5d	Release v2.3.71	2025-12-22 11:59:12 +00:00
Mervin Praison	8c735a60b2	Release v2.3.70	2025-12-22 10:43:48 +00:00
Mervin Praison	2f5f2f1c13	Release v2.3.69	2025-12-22 10:40:53 +00:00
Mervin Praison	0253073c68	Release v2.3.68	2025-12-22 10:35:45 +00:00
Mervin Praison	7f0e334e8c	Release v2.3.67	2025-12-22 09:14:33 +00:00
Mervin Praison	cacf1097ef	Release v2.3.66	2025-12-22 08:50:02 +00:00
Mervin Praison	09f01919d2	Fix duplicate --max-retries argument conflict in CLI parser	2025-12-21 05:19:08 +00:00
Mervin Praison	92397615f8	Release v2.3.65	2025-12-20 21:49:04 +00:00
Mervin Praison	4b21943a53	Release v2.3.64	2025-12-20 19:57:39 +00:00
Mervin Praison	af56a753bd	Release v2.3.63	2025-12-20 12:02:44 +00:00
Mervin Praison	15c2559a3c	Release v2.3.62	2025-12-20 11:59:46 +00:00
Mervin Praison	3215bdc0a6	Release v2.3.61	2025-12-20 11:15:22 +00:00
Mervin Praison	0ff0d83744	fix: Preserve real API keys for integration tests in conftest.py - Check for @pytest.mark.integration marker - Check if test file is in integration/ directory - This prevents conftest from overriding real API keys with 'test-key'	2025-12-20 03:58:13 +00:00
Mervin Praison	8f956f91dd	fix: Improve API key validation in integration test skip conditions - Check for 'sk-' prefix and minimum length (40 chars) to detect invalid keys - This ensures tests are skipped when using placeholder/test API keys in CI	2025-12-20 03:54:58 +00:00
Mervin Praison	954f6ac33d	fix: Skip integration tests when using test API key - test_auto_generator_integration.py: Skip when OPENAI_API_KEY contains 'test' - test_model_routing_real.py: Skip test_real_openai_call when using test key - test_cli_integrations.py: Skip all tests when using test key These tests require real API credentials and should not run in CI when only a placeholder test key is available.	2025-12-20 03:51:22 +00:00
Mervin Praison	264d11d8d9	fix: Fix failing unit tests for ClaudeCodeSDK and QueryRewriter CLI - test_use_sdk_option: Updated test to match actual behavior where use_sdk is only True when SDK is available (use_sdk and CLAUDE_SDK_AVAILABLE) - test_applies_rewrite_before_agent: Fixed mocking to properly mock _run_with_status_display and praisonaiagents submodules	2025-12-20 03:47:28 +00:00
Mervin Praison	88da2bdd7b	Release v2.3.60	2025-12-20 03:35:28 +00:00
Mervin Praison	b84e6c2460	Release v2.3.59	2025-12-20 03:28:48 +00:00
Mervin Praison	0434b23ada	Release v2.3.58	2025-12-20 03:08:31 +00:00
Mervin Praison	a9951e0694	Release v2.3.57	2025-12-19 16:47:09 +00:00
Mervin Praison	073062cd96	Release v2.3.56	2025-12-19 16:14:58 +00:00
Mervin Praison	936b422876	Release v2.3.55	2025-12-19 16:05:38 +00:00
Mervin Praison	99ab421ce2	Release v2.3.54	2025-12-19 16:04:43 +00:00
Mervin Praison	c1f334905e	Release v2.3.53	2025-12-19 15:38:30 +00:00
Mervin Praison	6d3ba7cdf8	Release v2.3.52	2025-12-19 15:34:19 +00:00
Mervin Praison	96616498d2	Release v2.3.51	2025-12-19 15:25:30 +00:00
Mervin Praison	1cad345415	Release v2.3.50	2025-12-19 15:19:29 +00:00
Mervin Praison	802e7e52f9	fix(readme): correct CLI docs URLs to include /docs/ prefix Changed all https://docs.praison.ai/cli/* URLs to https://docs.praison.ai/docs/cli/* Fixed 39 URLs total	2025-12-19 13:01:53 +00:00
Mervin Praison	53704f234c	Release v2.3.49	2025-12-19 12:49:12 +00:00
Mervin Praison	b575994d91	docs(readme): update AI commit CLI with security scanning and auto mode - Add full auto mode (-a) with stage, security check, commit, push - Document security scanning features - Add --no-verify flag documentation - List detected sensitive patterns	2025-12-19 12:35:49 +00:00
Mervin Praison	7ed8596261	Release v2.3.48	2025-12-19 12:35:02 +00:00
Mervin Praison	deef65c4bb	Release v2.3.47	2025-12-19 12:23:05 +00:00
Mervin Praison	d51327effd	Release v2.3.46	2025-12-19 12:22:29 +00:00
Mervin Praison	30ce8857e8	fix: suppress crewai.cli.config INFO log at startup	2025-12-19 12:18:23 +00:00
Mervin Praison	397874ccfc	Release v2.3.45	2025-12-19 12:16:12 +00:00
Mervin Praison	2016902e9a	docs: Add examples README with quick links to docs	2025-12-19 11:14:24 +00:00
Mervin Praison	d2d19511da	docs: Update README with Model Router info and fix links - Rename 'Router Agent' to 'Model Router' for consistency - Fix docs link to point to /features/model-router - Add model-routing template CLI command - Add link to Model Router documentation	2025-12-19 11:13:48 +00:00
Mervin Praison	bc9579b497	Add Code category with individual examples for external CLI integrations - Add Code category in Key Features table (like Workflows) - Add individual examples for each external agent: - claude_code_example.py - gemini_cli_example.py - codex_cli_example.py - cursor_cli_example.py - external_agents_example.py (all integrations) - code_editing_example.py - Link to individual docs pages for each integration - Organize code-related features under Code category	2025-12-19 11:07:25 +00:00
Mervin Praison	3161389341	Add External Agents CLI documentation to README - Add External Agents feature to Key Features table - Update Code Editing docs link to /code/editing - Add External Agents CLI section with examples - Include Python API usage examples - Add environment variables documentation - Link to detailed docs at docs.praison.ai/code/external-agents	2025-12-19 11:01:24 +00:00
Mervin Praison	b01eb2b8fa	Release v2.3.44	2025-12-19 10:50:33 +00:00
Mervin Praison	e609fb7425	Release v2.3.43	2025-12-19 10:32:33 +00:00
Mervin Praison	168ef98473	feat: Implement all remaining AutoAgents improvements (TDD) TODO 2: Single-Agent Option - Add SingleAgentStructure Pydantic model - Add single_agent parameter to AutoGenerator and WorkflowAutoGenerator TODO 3: LLM-Based Pattern Recommendation - Add PatternRecommendation Pydantic model (pattern, reasoning, confidence) - Add recommend_pattern_llm() method to WorkflowAutoGenerator TODO 4: Validation Gates - Add ValidationGate Pydantic model (criteria, pass_action, fail_action) - Add optional gates field to WorkflowStructure TODO 6: Pattern Support for AutoGenerator - Add pattern parameter to AutoGenerator - Add recommend_pattern() method to AutoGenerator - Add pattern-specific guidance in get_user_content() TODO 7: Merge Support for WorkflowAutoGenerator - Add merge parameter to generate() method - Add merge_with_existing_workflow() method TODO 10: Framework Support for WorkflowAutoGenerator - Add framework parameter to WorkflowAutoGenerator - Default framework is 'praisonai' Test Coverage: - 51 tests pass (up from 34) - 17 new tests added using TDD approach - All tests verify new functionality	2025-12-19 10:29:40 +00:00
Mervin Praison	ce2c2926f9	refactor: Add BaseAutoGenerator class and improve WorkflowAutoGenerator - Create BaseAutoGenerator with shared functionality (DRY principle) - Lazy-loaded instructor client (LiteLLM or OpenAI fallback) - Environment variable handling for model/API configuration - analyze_complexity() static method for task complexity detection - get_available_tools() static method for tool list - AVAILABLE_TOOLS constant shared between generators - Update AutoGenerator to inherit from BaseAutoGenerator - Removes ~40 lines of duplicate code - Maintains all existing functionality - Update WorkflowAutoGenerator to inherit from BaseAutoGenerator - Removes ~40 lines of duplicate code - Add dynamic agent count based on task complexity - Add tools list to workflow prompts - Improved prompt with 3-step analysis - Add 8 new unit tests for new functionality - TestBaseAutoGenerator: complexity analysis, tools list - TestWorkflowDynamicAgentCount: prompt improvements All 34 tests pass with no performance impact.	2025-12-19 10:23:29 +00:00
Mervin Praison	0a569e2b46	docs: Add AutoAgents workflow patterns examples and update README - Add autoagents_workflow_patterns.py with all 6 workflow patterns - Update auto_agents_example.py with new features and CLI examples - Add orchestrator-workers and evaluator-optimizer patterns to README CLI section - Update pattern options table with all 6 patterns - Add documentation links to examples	2025-12-19 09:54:05 +00:00
Mervin Praison	054aea8f66	Add MCP Registry badge and ignore MCP tokens	2025-12-18 22:18:33 +00:00
Mervin Praison	34f1293cfc	Release v2.3.42	2025-12-18 22:13:08 +00:00
Mervin Praison	7c1e32afc0	upate	2025-12-18 22:11:02 +00:00
				`@@ -0,0 +1 @@`
				`"""Unit tests for PraisonAI Agents evaluation framework."""`