Release v2.3.24

README.md

@@ -40,7 +40,17 @@ PraisonAI is a production-ready Multi-AI Agents framework with self-reflection,
 | 🔄 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) |
-| 🎭 AI Agent Workflow (Context Passing, Per-Step Agents) | [Example](examples/python/stateful/workflow-state-example.py) | [📖](https://docs.praison.ai/features/workflows) |
+| **🔄 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) |
+| ↳ Agentic Routing (`route()`) | [Example](examples/python/workflows/workflow_routing.py) | [📖](https://docs.praison.ai/features/routing) |
+| ↳ Parallel Execution (`parallel()`) | [Example](examples/python/workflows/workflow_parallel.py) | [📖](https://docs.praison.ai/features/parallelisation) |
+| ↳ Loop over List/CSV (`loop()`) | [Example](examples/python/workflows/workflow_loop_csv.py) | [📖](https://docs.praison.ai/features/repetitive) |
+| ↳ Evaluator-Optimizer (`repeat()`) | [Example](examples/python/workflows/workflow_repeat.py) | [📖](https://docs.praison.ai/features/evaluator-optimiser) |
+| ↳ Conditional Steps | [Example](examples/python/workflows/workflow_conditional.py) | [📖](https://docs.praison.ai/features/workflows) |
+| ↳ 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) |
@@ -103,20 +113,6 @@ PraisonAI is a production-ready Multi-AI Agents framework with self-reflection,
 | @ @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) |
-| ⏸️ Workflow Checkpoints | [Example](examples/python/workflows/workflow_checkpoints.py) | [📖](https://docs.praison.ai/cli/session) |
-| 🔀 Workflow Branching | [Example](examples/python/workflows/workflow_branching.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 🔁 Workflow Loops | [Example](examples/python/workflows/workflow_loops.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 📞 Task Callbacks | [Example](examples/python/workflows/task_callbacks.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 🛑 Workflow Early Stop | [Example](examples/python/workflows/workflow_early_stop.py) | [📖](https://docs.praison.ai/features/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) |
-| 🔄 Conditional Steps | [Example](examples/python/workflows/workflow_conditional.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 🎭 Mixed Steps | [Example](examples/python/workflows/workflow_mixed_steps.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 🔀 Agentic Routing | [Example](examples/python/workflows/workflow_routing.py) | [📖](https://docs.praison.ai/features/workflow-routing) |
-| ⚡ Parallel Execution | [Example](examples/python/workflows/workflow_parallel.py) | [📖](https://docs.praison.ai/features/workflow-parallel) |
-| 🔁 Loop over List | [Example](examples/python/workflows/workflow_loop_list.py) | [📖](https://docs.praison.ai/features/workflow-loop) |
-| 📄 Loop over CSV | [Example](examples/python/workflows/workflow_loop_csv.py) | [📖](https://docs.praison.ai/features/workflow-loop) |
-| 🔄 Evaluator-Optimizer | [Example](examples/python/workflows/workflow_repeat.py) | [📖](https://docs.praison.ai/features/workflow-repeat) |
 
 ## Supported Providers
 
@@ -467,8 +463,8 @@ result = asyncio.run(manager.aexecute("deploy", default_llm="gpt-4o-mini"))
 |----------|-------------|
 | Simple function pipelines | `Workflow` class ⭐ |
 | Agent-only pipelines | `Workflow` class |
-| CSV batch processing | `process="workflow"` |
-| Complex task routing | `process="workflow"` |
+| CSV batch processing | `Workflow` + `loop()` |
+| Complex task routing | `Workflow` + `route()` |
 | Markdown templates | `WorkflowManager` |
 | Early stop / conditional | `Workflow` class |
 
@@ -562,35 +558,178 @@ workflow = Workflow(steps=[
         max_iterations=5
     )
 ])
+
+# 5. CALLBACKS - Monitor workflow execution
+workflow = Workflow(
+    steps=[step1, step2],
+    on_workflow_start=lambda w, i: print(f"Starting: {i}"),
+    on_step_complete=lambda name, r: print(f"{name}: {r.output[:50]}"),
+    on_workflow_complete=lambda w, r: print(f"Done: {r['status']}")
+)
+
+# 6. GUARDRAILS - Validate and retry
+def validate(result):
+    return ("error" not in result.output, "Fix the error")
+
+workflow = Workflow(steps=[
+    WorkflowStep(name="gen", handler=generator, guardrail=validate, max_retries=3)
+])
 ```
 
-### Task Callbacks & Variables (process="workflow")
+### Workflow with Agents
+
+Use Agent objects directly as workflow steps:
 
 ```python
-from praisonaiagents import Agent, Task, PraisonAIAgents
+from praisonaiagents import Agent, Workflow
+from praisonaiagents.workflows import route, parallel
 
-# Per-task agent config (no need to create agent separately)
-task = Task(
-    description="Research {{topic}}",
-    agent_config={"name": "Researcher", "role": "Expert", "llm": "gpt-4o-mini"},
-    variables={"topic": "AI trends"}
+# 1. SEQUENTIAL AGENTS
+researcher = Agent(name="Researcher", role="Research expert", tools=[tavily_search])
+writer = Agent(name="Writer", role="Content writer")
+editor = Agent(name="Editor", role="Editor")
+
+workflow = Workflow(steps=[researcher, writer, editor])
+result = workflow.start("Research and write about AI")
+
+# 2. PARALLEL AGENTS
+workflow = Workflow(steps=[
+    parallel([researcher1, researcher2, researcher3]),
+    aggregator_agent
+])
+
+# 3. ROUTE TO AGENTS
+workflow = Workflow(steps=[
+    classifier_function,
+    route({
+        "technical": [tech_agent],
+        "creative": [creative_agent],
+        "default": [general_agent]
+    })
+])
+
+# 4. WITH PLANNING & REASONING
+workflow = Workflow(
+    steps=[researcher, writer, editor],
+    planning=True,           # Create execution plan
+    planning_llm="gpt-4o",   # LLM for planning
+    reasoning=True,          # Chain-of-thought reasoning
+    verbose=True
 )
 
-# Callbacks for workflow execution
-def on_start(task, task_id):
-    print(f"Starting: {task.name}")
+# 5. TOOLS PER STEP
+workflow = Workflow(steps=[
+    WorkflowStep(
+        name="research",
+        action="Research {{topic}}",
+        tools=[tavily_search, web_scraper],
+        agent_config={"name": "Researcher", "role": "Expert"}
+    )
+])
 
-def on_complete(task, output):
-    print(f"Completed: {task.name}")
+# 6. OUTPUT TO FILE / IMAGES / PYDANTIC
+from pydantic import BaseModel
 
-agents = PraisonAIAgents(
-    agents=[agent],
-    tasks=[task],
-    process="workflow",
-    on_task_start=on_start,
-    on_task_complete=on_complete,
-    variables={"global_var": "shared_value"}  # Global variables for all tasks
+class Report(BaseModel):
+    title: str
+    content: str
+
+workflow = Workflow(steps=[
+    WorkflowStep(name="analyze", action="Analyze image", images=["image.jpg"]),
+    WorkflowStep(name="report", action="Generate report", output_pydantic=Report),
+    WorkflowStep(name="save", action="Save results", output_file="output/report.txt")
+])
+
+# 7. ASYNC EXECUTION
+import asyncio
+
+async def main():
+    result = await workflow.astart("input")
+    print(result)
+
+asyncio.run(main())
+
+# 8. STATUS TRACKING
+workflow.status  # "not_started" | "running" | "completed"
+workflow.step_statuses  # {"step1": "completed", "step2": "skipped"}
+
+# 9. MEMORY CONFIG
+workflow = Workflow(
+    steps=[researcher, writer],
+    memory_config={"provider": "chroma", "persist": True, "collection": "my_workflow"}
 )
+result1 = workflow.start("Research AI")
+result2 = workflow.start("Continue the research")  # Remembers first run
+```
+
+### YAML Workflow Template
+
+```yaml
+# .praison/workflows/research.yaml
+name: Research Workflow
+description: Research and write content with all patterns
+
+agents:
+  researcher:
+    role: Research Expert
+    goal: Find accurate information
+    tools: [tavily_search, web_scraper]
+  writer:
+    role: Content Writer
+    goal: Write engaging content
+  editor:
+    role: Editor
+    goal: Polish content
+
+steps:
+  # Sequential
+  - agent: researcher
+    action: Research {{topic}}
+    output_variable: research_data
+
+  # Routing
+  - name: classifier
+    action: Classify content type
+    route:
+      technical: [tech_handler]
+      creative: [creative_handler]
+      default: [general_handler]
+
+  # Parallel
+  - name: parallel_research
+    parallel:
+      - agent: researcher
+        action: Research market
+      - agent: researcher
+        action: Research competitors
+
+  # Loop
+  - agent: writer
+    action: Write about {{item}}
+    loop_over: topics
+    loop_var: item
+
+  # Repeat (evaluator-optimizer)
+  - agent: editor
+    action: Review and improve
+    repeat:
+      until: "quality > 8"
+      max_iterations: 3
+
+  # Output to file
+  - agent: writer
+    action: Write final report
+    output_file: output/{{topic}}_report.md
+
+variables:
+  topic: AI trends
+  topics: [ML, NLP, Vision]
+
+planning: true
+planning_llm: gpt-4o
+memory_config:
+  provider: chroma
+  persist: true
 ```
 
 ### 9. Hooks
@@ -736,6 +875,12 @@ praisonai workflow run deploy --workflow-var environment=staging --workflow-var
 # Execute with planning mode (AI creates sub-steps for each workflow step)
 praisonai workflow run "Research Blog" --planning --verbose
 
+# Execute with reasoning mode (chain-of-thought)
+praisonai workflow run "Analysis" --reasoning --verbose
+
+# Execute with memory enabled
+praisonai workflow run "Research" --memory
+
 # Show workflow details
 praisonai workflow show deploy
 
@@ -747,8 +892,23 @@ praisonai "What is AI?" --workflow "Research,Summarize" --save
 
 # Inline workflow with step actions
 praisonai "GPT-5" --workflow "Research:Search for info,Write:Write blog" --tools tavily
+
+# Workflow CLI help
+praisonai workflow help
 ```
 
+**Workflow CLI Options:**
+| Flag | Description |
+|------|-------------|
+| `--workflow-var key=value` | Set workflow variable (can be repeated) |
+| `--llm <model>` | LLM model (e.g., openai/gpt-4o-mini) |
+| `--tools <tools>` | Tools (comma-separated, e.g., tavily) |
+| `--planning` | Enable planning mode |
+| `--reasoning` | Enable reasoning mode |
+| `--memory` | Enable memory |
+| `--verbose` | Enable verbose output |
+| `--save` | Save output to file |
+
 ### Hooks CLI:
 ```bash
 # List configured hooks

docker/Dockerfile

@@ -16,7 +16,7 @@ RUN mkdir -p /root/.praison
 # Install Python packages (using latest versions)
 RUN pip install --no-cache-dir \
     flask \
-    "praisonai>=2.3.21" \
+    "praisonai>=2.3.24" \
     "praisonai[api]" \
     gunicorn \
     markdown

docker/Dockerfile.chat

@@ -16,7 +16,7 @@ RUN mkdir -p /root/.praison
 # Install Python packages (using latest versions)
 RUN pip install --no-cache-dir \
     praisonai_tools \
-    "praisonai>=2.3.21" \
+    "praisonai>=2.3.24" \
     "praisonai[chat]" \
     "embedchain[github,youtube]"
 

docker/Dockerfile.dev

@@ -20,7 +20,7 @@ RUN mkdir -p /root/.praison
 # Install Python packages (using latest versions)
 RUN pip install --no-cache-dir \
     praisonai_tools \
-    "praisonai>=2.3.21" \
+    "praisonai>=2.3.24" \
     "praisonai[ui]" \
     "praisonai[chat]" \
     "praisonai[realtime]" \

docker/Dockerfile.ui

@@ -16,7 +16,7 @@ RUN mkdir -p /root/.praison
 # Install Python packages (using latest versions)
 RUN pip install --no-cache-dir \
     praisonai_tools \
-    "praisonai>=2.3.21" \
+    "praisonai>=2.3.24" \
     "praisonai[ui]" \
     "praisonai[crewai]"
 

src/praisonai-agents/praisonaiagents/process/process.py

@@ -826,7 +826,34 @@ Provide a JSON with the structure:
         logging.info("Hierarchical task execution finished")
 
     def workflow(self):
-        """Synchronous version of workflow method"""
+        """Synchronous version of workflow method.
+        
+        DEPRECATED: This method is deprecated. Use the Workflow class instead:
+        
+        ```python
+        from praisonaiagents import Workflow, StepResult
+        from praisonaiagents.workflows import route, parallel, loop, repeat
+        
+        workflow = Workflow(steps=[step1, step2])
+        result = workflow.start("input")
+        ```
+        
+        The Workflow class provides a simpler API with more features:
+        - route() for decision-based branching
+        - parallel() for concurrent execution
+        - loop() for iteration over lists/CSV
+        - repeat() for evaluator-optimizer patterns
+        - Callbacks (on_step_start, on_step_complete, etc.)
+        - Guardrails with validation feedback
+        - Status tracking
+        """
+        import warnings
+        warnings.warn(
+            "process='workflow' is deprecated. Use the Workflow class instead: "
+            "from praisonaiagents import Workflow; workflow = Workflow(steps=[...]); workflow.start()",
+            DeprecationWarning,
+            stacklevel=2
+        )
         current_iter = 0  # Track how many times we've looped
         # Build workflow relationships first
         for task in self.tasks.values():

src/praisonai-agents/praisonaiagents/workflows/workflows.py

@@ -23,7 +23,7 @@ import os
 import re
 import logging
 from pathlib import Path
-from typing import Any, Dict, List, Optional, Callable, Literal
+from typing import Any, Dict, List, Optional, Callable, Literal, Tuple
 from dataclasses import dataclass, field
 
 logger = logging.getLogger(__name__)
@@ -190,7 +190,7 @@ class WorkflowStep:
     should_run: Optional[Callable[['WorkflowContext'], bool]] = None  # Function to check if step should run
     handler: Optional[Callable[['WorkflowContext'], 'StepResult']] = None  # Custom function instead of agent
     on_error: Literal["stop", "continue", "retry"] = "stop"
-    max_retries: int = 1
+    max_retries: int = 3
     
     # Context passing fields
     context_from: Optional[List[str]] = None  # Which previous steps to include context from
@@ -210,6 +210,26 @@ class WorkflowStep:
     loop_over: Optional[str] = None  # Variable name to iterate over (e.g., "items")
     loop_var: str = "item"  # Variable name for current item in loop
     
+    # Guardrail (like process="workflow")
+    guardrail: Optional[Callable[['StepResult'], Tuple[bool, Any]]] = None  # Returns (is_valid, feedback)
+    
+    # Output configuration (migrated from Task)
+    output_file: Optional[str] = None  # Save output to file
+    output_json: Optional[Any] = None  # Pydantic model for JSON output
+    output_pydantic: Optional[Any] = None  # Pydantic model for structured output
+    
+    # Image handling (migrated from Task)
+    images: Optional[List[str]] = None  # Image paths/URLs for vision tasks
+    
+    # Execution control (migrated from Task)
+    async_execution: bool = False  # Run step asynchronously
+    quality_check: bool = True  # Enable quality validation
+    rerun: bool = True  # Allow step to be rerun
+    
+    # Status tracking
+    status: str = "pending"  # pending, running, completed, failed, skipped
+    retry_count: int = 0
+    
     # Backward compatibility aliases
     @property
     def evaluator(self):
@@ -238,7 +258,14 @@ class WorkflowStep:
             "next_steps": self.next_steps,
             "branch_condition": self.branch_condition,
             "loop_over": self.loop_over,
-            "loop_var": self.loop_var
+            "loop_var": self.loop_var,
+            "output_file": self.output_file,
+            "output_json": self.output_json is not None,
+            "output_pydantic": self.output_pydantic is not None,
+            "images": self.images,
+            "async_execution": self.async_execution,
+            "quality_check": self.quality_check,
+            "rerun": self.rerun
         }
 
 
@@ -257,6 +284,19 @@ class Workflow:
     memory_config: Optional[Dict[str, Any]] = None  # Memory configuration
     planning: bool = False  # Enable planning mode
     planning_llm: Optional[str] = None  # LLM for planning
+    reasoning: bool = False  # Enable chain-of-thought reasoning
+    verbose: bool = False  # Enable verbose output
+    
+    # Callbacks (like process="workflow")
+    on_workflow_start: Optional[Callable[['Workflow', str], None]] = None  # (workflow, input)
+    on_workflow_complete: Optional[Callable[['Workflow', Dict[str, Any]], None]] = None  # (workflow, result)
+    on_step_start: Optional[Callable[[str, 'WorkflowContext'], None]] = None  # (step_name, context)
+    on_step_complete: Optional[Callable[[str, 'StepResult'], None]] = None  # (step_name, result)
+    on_step_error: Optional[Callable[[str, Exception], None]] = None  # (step_name, error)
+    
+    # Status tracking
+    status: str = "not_started"  # not_started, running, completed, failed
+    step_statuses: Dict[str, str] = field(default_factory=dict)  # {step_name: status}
     
     def to_dict(self) -> Dict[str, Any]:
         return {
@@ -307,12 +347,32 @@ class Workflow:
         # Use default LLM if not specified
         model = llm or self.default_llm or "gpt-4o-mini"
         
+        # Use workflow verbose setting if not overridden
+        verbose = verbose or self.verbose
+        
         # Add input to variables
         all_variables = {**self.variables, "input": input}
         
         results = []
         previous_output = None
         
+        # Update workflow status
+        self.status = "running"
+        
+        # Call on_workflow_start callback
+        if self.on_workflow_start:
+            try:
+                self.on_workflow_start(self, input)
+            except Exception as e:
+                logger.error(f"on_workflow_start callback failed: {e}")
+        
+        # Planning mode - create execution plan before running
+        if self.planning:
+            plan = self._create_plan(input, model, verbose)
+            if plan and verbose:
+                print(f"📋 Execution Plan: {plan}")
+            all_variables["execution_plan"] = plan
+        
         # Process steps (may include Route, Parallel, Loop, Repeat)
         i = 0
         while i < len(self.steps):
@@ -374,68 +434,197 @@ class Workflow:
                 variables=all_variables.copy()
             )
             
+            # Update step status
+            if hasattr(step, 'status'):
+                step.status = "running"
+            self.step_statuses[step.name] = "running"
+            
+            # Call on_step_start callback
+            if self.on_step_start:
+                try:
+                    self.on_step_start(step.name, context)
+                except Exception as e:
+                    logger.error(f"on_step_start callback failed: {e}")
+            
             # Check should_run condition
             if step.should_run:
                 try:
                     if not step.should_run(context):
                         if verbose:
                             print(f"⏭️ Skipped: {step.name}")
+                        if hasattr(step, 'status'):
+                            step.status = "skipped"
+                        self.step_statuses[step.name] = "skipped"
                         i += 1
                         continue
                 except Exception as e:
                     logger.error(f"should_run failed for {step.name}: {e}")
             
-            # Execute step
+            # Execute step with retry and guardrail support
             output = None
             stop = False
+            step_error = None
+            max_retries = getattr(step, 'max_retries', 3)
+            retry_count = 0
+            validation_feedback = None
             
-            if step.handler:
-                # Custom handler function
+            while retry_count <= max_retries:
+                step_error = None
+                
+                # Add validation feedback to context if retrying
+                if validation_feedback:
+                    context.variables["validation_feedback"] = validation_feedback
+                
                 try:
-                    result = step.handler(context)
-                    if isinstance(result, StepResult):
-                        output = result.output
-                        stop = result.stop_workflow
-                        if result.variables:
-                            all_variables.update(result.variables)
-                    else:
-                        output = str(result)
+                    if step.handler:
+                        # Custom handler function
+                        result = step.handler(context)
+                        if isinstance(result, StepResult):
+                            output = result.output
+                            stop = result.stop_workflow
+                            if result.variables:
+                                all_variables.update(result.variables)
+                        else:
+                            output = str(result)
+                            
+                    elif step.agent:
+                        # Direct agent with tools
+                        # Substitute variables in action
+                        action = step.action or input
+                        for key, value in all_variables.items():
+                            action = action.replace(f"{{{{{key}}}}}", str(value))
+                        if previous_output:
+                            action = action.replace("{{previous_output}}", str(previous_output))
+                        action = action.replace("{{input}}", input)
+                        
+                        # Add reasoning prompt if enabled
+                        if self.reasoning:
+                            action = f"Think step by step and reason through this task:\n\n{action}"
+                        
+                        # Add validation feedback if retrying
+                        if validation_feedback:
+                            action = f"{action}\n\nPrevious attempt feedback: {validation_feedback}"
+                        
+                        # Handle images if present
+                        step_images = getattr(step, 'images', None)
+                        if step_images:
+                            output = step.agent.chat(action, images=step_images)
+                        else:
+                            output = step.agent.chat(action)
+                        
+                        # Handle output_pydantic if present
+                        output_pydantic = getattr(step, 'output_pydantic', None)
+                        if output_pydantic and output:
+                            try:
+                                # Try to parse output as Pydantic model
+                                if hasattr(output_pydantic, 'model_validate_json'):
+                                    parsed = output_pydantic.model_validate_json(output)
+                                    output = parsed.model_dump_json()
+                            except Exception as e:
+                                logger.debug(f"Pydantic parsing failed: {e}")
+                            
+                    elif step.action:
+                        # Action with agent_config - create temporary agent
+                        from ..agent.agent import Agent
+                        config = step.agent_config or self.default_agent_config or {}
+                        
+                        # Get tools from step or config
+                        step_tools = step.tools or config.get("tools", [])
+                        
+                        temp_agent = Agent(
+                            name=config.get("name", step.name),
+                            role=config.get("role", "Assistant"),
+                            goal=config.get("goal", "Complete the task"),
+                            llm=config.get("llm", model),
+                            tools=step_tools if step_tools else None,
+                            verbose=verbose,
+                            reasoning=self.reasoning
+                        )
+                        # Substitute variables in action
+                        action = step.action
+                        for key, value in all_variables.items():
+                            action = action.replace(f"{{{{{key}}}}}", str(value))
+                        if previous_output:
+                            action = action.replace("{{previous_output}}", str(previous_output))
+                        action = action.replace("{{input}}", input)
+                        
+                        # Add reasoning prompt if enabled
+                        if self.reasoning:
+                            action = f"Think step by step and reason through this task:\n\n{action}"
+                        
+                        if validation_feedback:
+                            action = f"{action}\n\nPrevious attempt feedback: {validation_feedback}"
+                        output = temp_agent.chat(action)
+                        
                 except Exception as e:
+                    step_error = e
                     output = f"Error: {e}"
-                    
-            elif step.agent:
-                # Direct agent
+                    if self.on_step_error:
+                        try:
+                            self.on_step_error(step.name, e)
+                        except Exception:
+                            pass
+                
+                # Check guardrail if present
+                guardrail = getattr(step, 'guardrail', None)
+                if guardrail and output and not step_error:
+                    try:
+                        is_valid, feedback = guardrail(StepResult(output=output))
+                        if not is_valid:
+                            validation_feedback = str(feedback)
+                            retry_count += 1
+                            if verbose:
+                                print(f"⚠️ {step.name} failed validation (attempt {retry_count}/{max_retries}): {feedback}")
+                            continue  # Retry
+                    except Exception as e:
+                        logger.error(f"Guardrail failed for {step.name}: {e}")
+                
+                # Success - break out of retry loop
+                break
+            
+            # Update step status
+            if step_error:
+                if hasattr(step, 'status'):
+                    step.status = "failed"
+                self.step_statuses[step.name] = "failed"
+            else:
+                if hasattr(step, 'status'):
+                    step.status = "completed"
+                self.step_statuses[step.name] = "completed"
+            
+            # Create step result for callback
+            step_result = StepResult(output=output or "", stop_workflow=stop)
+            
+            # Call on_step_complete callback
+            if self.on_step_complete:
                 try:
-                    output = step.agent.chat(step.action or input)
+                    self.on_step_complete(step.name, step_result)
                 except Exception as e:
-                    output = f"Error: {e}"
-                    
-            elif step.action:
-                # Action with agent_config - create temporary agent
+                    logger.error(f"on_step_complete callback failed: {e}")
+            
+            # Handle output_file - save output to file
+            if hasattr(step, 'output_file') and step.output_file and output:
                 try:
-                    from ..agent.agent import Agent
-                    config = step.agent_config or self.default_agent_config or {}
-                    temp_agent = Agent(
-                        name=config.get("name", step.name),
-                        role=config.get("role", "Assistant"),
-                        goal=config.get("goal", "Complete the task"),
-                        llm=config.get("llm", model),
-                        verbose=verbose
-                    )
-                    # Substitute variables in action
-                    action = step.action
+                    import os
+                    output_path = step.output_file
+                    # Substitute variables in path
                     for key, value in all_variables.items():
-                        action = action.replace(f"{{{{{key}}}}}", str(value))
-                    if previous_output:
-                        action = action.replace("{{previous_output}}", str(previous_output))
-                    output = temp_agent.chat(action)
+                        output_path = output_path.replace(f"{{{{{key}}}}}", str(value))
+                    # Create directory if needed
+                    os.makedirs(os.path.dirname(output_path) if os.path.dirname(output_path) else ".", exist_ok=True)
+                    with open(output_path, "w") as f:
+                        f.write(str(output))
+                    if verbose:
+                        print(f"📁 Saved output to: {output_path}")
                 except Exception as e:
-                    output = f"Error: {e}"
+                    logger.error(f"Failed to save output to file: {e}")
             
             # Store result
             results.append({
                 "step": step.name,
-                "output": output
+                "output": output,
+                "status": self.step_statuses.get(step.name, "completed"),
+                "retries": retry_count
             })
             previous_output = output
             
@@ -454,11 +643,59 @@ class Workflow:
             
             i += 1
         
-        return {
+        # Update workflow status
+        self.status = "completed"
+        
+        final_result = {
             "output": previous_output,
             "steps": results,
-            "variables": all_variables
+            "variables": all_variables,
+            "status": self.status
         }
+        
+        # Call on_workflow_complete callback
+        if self.on_workflow_complete:
+            try:
+                self.on_workflow_complete(self, final_result)
+            except Exception as e:
+                logger.error(f"on_workflow_complete callback failed: {e}")
+        
+        return final_result
+    
+    async def astart(
+        self,
+        input: str = "",
+        llm: Optional[str] = None,
+        verbose: bool = False
+    ) -> Dict[str, Any]:
+        """Async version of start() for async workflow execution.
+        
+        Args:
+            input: The input text/prompt for the workflow
+            llm: LLM model to use (default: gpt-4o-mini)
+            verbose: Print step outputs
+            
+        Returns:
+            Dict with 'output' (final result) and 'steps' (all step results)
+        """
+        import asyncio
+        
+        # Run the synchronous version in a thread pool
+        loop = asyncio.get_event_loop()
+        result = await loop.run_in_executor(
+            None,
+            lambda: self.run(input, llm, verbose)
+        )
+        return result
+    
+    async def arun(
+        self,
+        input: str = "",
+        llm: Optional[str] = None,
+        verbose: bool = False
+    ) -> Dict[str, Any]:
+        """Alias for astart() for backward compatibility."""
+        return await self.astart(input, llm, verbose)
     
     def _normalize_steps(self) -> List['WorkflowStep']:
         """Convert mixed steps (Agent, function, WorkflowStep) to WorkflowStep list."""
@@ -489,21 +726,89 @@ class Workflow:
         
         return normalized
     
+    def _create_plan(self, input: str, model: str, verbose: bool) -> Optional[str]:
+        """Create an execution plan for the workflow using LLM.
+        
+        Args:
+            input: The workflow input
+            model: LLM model to use
+            verbose: Print verbose output
+            
+        Returns:
+            Execution plan as string, or None if planning fails
+        """
+        try:
+            from ..agent.agent import Agent
+            
+            # Describe the steps
+            step_descriptions = []
+            for i, step in enumerate(self.steps):
+                if isinstance(step, WorkflowStep):
+                    step_descriptions.append(f"{i+1}. {step.name}: {step.description or step.action or 'Execute step'}")
+                elif hasattr(step, 'name'):
+                    role = getattr(step, 'role', 'Agent')
+                    step_descriptions.append(f"{i+1}. {step.name} ({role})")
+                elif callable(step):
+                    step_descriptions.append(f"{i+1}. {getattr(step, '__name__', 'function')}")
+                elif isinstance(step, (Route, Parallel, Loop, Repeat)):
+                    step_descriptions.append(f"{i+1}. {type(step).__name__} pattern")
+                else:
+                    step_descriptions.append(f"{i+1}. Step {i+1}")
+            
+            steps_text = "\n".join(step_descriptions)
+            
+            planning_prompt = f"""You are a workflow planner. Given the following workflow steps and input, create a brief execution plan.
+
+Workflow: {self.name}
+Description: {self.description}
+
+Steps:
+{steps_text}
+
+Input: {input}
+
+Create a brief execution plan (2-3 sentences) describing how to best accomplish this task using the available steps. Focus on the key objectives and any important considerations."""
+
+            planner = Agent(
+                name="Planner",
+                role="Workflow Planner",
+                goal="Create efficient execution plans",
+                llm=self.planning_llm or model,
+                verbose=False
+            )
+            
+            plan = planner.chat(planning_prompt)
+            return plan
+            
+        except Exception as e:
+            logger.error(f"Planning failed: {e}")
+            return None
+    
     def _normalize_single_step(self, step: Any, index: int) -> 'WorkflowStep':
-        """Normalize a single step to WorkflowStep."""
+        """Normalize a single step to WorkflowStep.
+        
+        Supports:
+        - WorkflowStep objects (passed through)
+        - Agent objects (wrapped with agent reference and tools)
+        - Callable functions (wrapped as handler)
+        - Strings (used as action)
+        """
         if isinstance(step, WorkflowStep):
             return step
+        elif hasattr(step, 'chat'):
+            # It's an Agent - wrap with agent reference and preserve tools
+            agent_tools = getattr(step, 'tools', None)
+            return WorkflowStep(
+                name=getattr(step, 'name', f'agent_{index+1}'),
+                agent=step,
+                tools=agent_tools,
+                action="{{previous_output}}" if index > 0 else "{{input}}"
+            )
         elif callable(step):
             return WorkflowStep(
                 name=getattr(step, '__name__', f'step_{index+1}'),
                 handler=step
             )
-        elif hasattr(step, 'chat'):
-            return WorkflowStep(
-                name=getattr(step, 'name', f'agent_{index+1}'),
-                agent=step,
-                action="{{input}}"
-            )
         else:
             return WorkflowStep(
                 name=f'step_{index+1}',
@@ -922,6 +1227,13 @@ class WorkflowManager:
         loop_var_pattern = re.compile(r'loop_var:\s*(\w+)', re.IGNORECASE)
         branch_pattern = re.compile(r'```branch\s*\n(.*?)\n```', re.DOTALL)
         
+        # New patterns for route, parallel, images, output_file
+        route_pattern = re.compile(r'```route\s*\n(.*?)\n```', re.DOTALL)
+        parallel_pattern = re.compile(r'```parallel\s*\n(.*?)\n```', re.DOTALL)
+        images_pattern = re.compile(r'```images\s*\n(.*?)\n```', re.DOTALL)
+        output_file_pattern = re.compile(r'output_file:\s*(.+)', re.IGNORECASE)
+        repeat_pattern = re.compile(r'```repeat\s*\n(.*?)\n```', re.DOTALL)
+        
         # Find all step headers
         matches = list(step_pattern.finditer(body))
         
@@ -997,6 +1309,41 @@ class WorkflowManager:
             if loop_var_match:
                 loop_var = loop_var_match.group(1)
             
+            # Extract route pattern (same as branch_condition)
+            route_match = route_pattern.search(step_content)
+            if route_match and not branch_condition:
+                branch_condition = self._parse_branch_condition(route_match.group(1).strip())
+            
+            # Extract parallel pattern
+            parallel_steps = None
+            parallel_match = parallel_pattern.search(step_content)
+            if parallel_match:
+                parallel_str = parallel_match.group(1).strip()
+                # Parse as list of step names
+                parallel_steps = [s.strip().lstrip('- ').strip('"\'') for s in parallel_str.split('\n') if s.strip()]
+                # Set next_steps to parallel steps if not already set
+                if not next_steps:
+                    next_steps = parallel_steps
+            
+            # Extract images
+            images = None
+            images_match = images_pattern.search(step_content)
+            if images_match:
+                images_str = images_match.group(1).strip()
+                images = [s.strip() for s in images_str.split('\n') if s.strip()]
+            
+            # Extract output_file
+            output_file = None
+            output_file_match = output_file_pattern.search(step_content)
+            if output_file_match:
+                output_file = output_file_match.group(1).strip()
+            
+            # Extract repeat config
+            repeat_config = None
+            repeat_match = repeat_pattern.search(step_content)
+            if repeat_match:
+                repeat_config = self._parse_agent_config(repeat_match.group(1).strip())
+            
             # Get description (text before action block)
             description = step_content
             if action_match:
@@ -1010,8 +1357,14 @@ class WorkflowManager:
             description = re.sub(r'next_steps:.*', '', description)
             description = re.sub(r'loop_over:.*', '', description)
             description = re.sub(r'loop_var:.*', '', description)
+            description = re.sub(r'output_file:.*', '', description)
             description = description.strip()
             
+            # Set max_retries from repeat config
+            max_retries = 3
+            if repeat_config and 'max_iterations' in repeat_config:
+                max_retries = int(repeat_config['max_iterations'])
+            
             steps.append(WorkflowStep(
                 name=step_name,
                 description=description,
@@ -1025,7 +1378,10 @@ class WorkflowManager:
                 next_steps=next_steps,
                 branch_condition=branch_condition,
                 loop_over=loop_over,
-                loop_var=loop_var
+                loop_var=loop_var,
+                output_file=output_file,
+                images=images,
+                max_retries=max_retries
             ))
         
         return steps

src/praisonai-agents/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "praisonaiagents"
-version = "0.0.199"
+version = "0.1.2"
 description = "Praison AI agents for completing complex tasks with Self Reflection Agents"
 requires-python = ">=3.10"
 authors = [

src/praisonai-agents/uv.lock

@@ -2422,7 +2422,7 @@ wheels = [
 
 [[package]]
 name = "praisonaiagents"
-version = "0.0.199"
+version = "0.1.2"
 source = { editable = "." }
 dependencies = [
     { name = "aiohttp" },

src/praisonai/README.md

@@ -40,7 +40,17 @@ PraisonAI is a production-ready Multi-AI Agents framework with self-reflection,
 | 🔄 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) |
-| 🎭 AI Agent Workflow (Context Passing, Per-Step Agents) | [Example](examples/python/stateful/workflow-state-example.py) | [📖](https://docs.praison.ai/features/workflows) |
+| **🔄 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) |
+| ↳ Agentic Routing (`route()`) | [Example](examples/python/workflows/workflow_routing.py) | [📖](https://docs.praison.ai/features/routing) |
+| ↳ Parallel Execution (`parallel()`) | [Example](examples/python/workflows/workflow_parallel.py) | [📖](https://docs.praison.ai/features/parallelisation) |
+| ↳ Loop over List/CSV (`loop()`) | [Example](examples/python/workflows/workflow_loop_csv.py) | [📖](https://docs.praison.ai/features/repetitive) |
+| ↳ Evaluator-Optimizer (`repeat()`) | [Example](examples/python/workflows/workflow_repeat.py) | [📖](https://docs.praison.ai/features/evaluator-optimiser) |
+| ↳ Conditional Steps | [Example](examples/python/workflows/workflow_conditional.py) | [📖](https://docs.praison.ai/features/workflows) |
+| ↳ 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) |
@@ -103,20 +113,6 @@ PraisonAI is a production-ready Multi-AI Agents framework with self-reflection,
 | @ @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) |
-| ⏸️ Workflow Checkpoints | [Example](examples/python/workflows/workflow_checkpoints.py) | [📖](https://docs.praison.ai/cli/session) |
-| 🔀 Workflow Branching | [Example](examples/python/workflows/workflow_branching.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 🔁 Workflow Loops | [Example](examples/python/workflows/workflow_loops.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 📞 Task Callbacks | [Example](examples/python/workflows/task_callbacks.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 🛑 Workflow Early Stop | [Example](examples/python/workflows/workflow_early_stop.py) | [📖](https://docs.praison.ai/features/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) |
-| 🔄 Conditional Steps | [Example](examples/python/workflows/workflow_conditional.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 🎭 Mixed Steps | [Example](examples/python/workflows/workflow_mixed_steps.py) | [📖](https://docs.praison.ai/features/workflows) |
-| 🔀 Agentic Routing | [Example](examples/python/workflows/workflow_routing.py) | [📖](https://docs.praison.ai/features/workflow-routing) |
-| ⚡ Parallel Execution | [Example](examples/python/workflows/workflow_parallel.py) | [📖](https://docs.praison.ai/features/workflow-parallel) |
-| 🔁 Loop over List | [Example](examples/python/workflows/workflow_loop_list.py) | [📖](https://docs.praison.ai/features/workflow-loop) |
-| 📄 Loop over CSV | [Example](examples/python/workflows/workflow_loop_csv.py) | [📖](https://docs.praison.ai/features/workflow-loop) |
-| 🔄 Evaluator-Optimizer | [Example](examples/python/workflows/workflow_repeat.py) | [📖](https://docs.praison.ai/features/workflow-repeat) |
 
 ## Supported Providers
 
@@ -467,8 +463,8 @@ result = asyncio.run(manager.aexecute("deploy", default_llm="gpt-4o-mini"))
 |----------|-------------|
 | Simple function pipelines | `Workflow` class ⭐ |
 | Agent-only pipelines | `Workflow` class |
-| CSV batch processing | `process="workflow"` |
-| Complex task routing | `process="workflow"` |
+| CSV batch processing | `Workflow` + `loop()` |
+| Complex task routing | `Workflow` + `route()` |
 | Markdown templates | `WorkflowManager` |
 | Early stop / conditional | `Workflow` class |
 
@@ -562,35 +558,178 @@ workflow = Workflow(steps=[
         max_iterations=5
     )
 ])
+
+# 5. CALLBACKS - Monitor workflow execution
+workflow = Workflow(
+    steps=[step1, step2],
+    on_workflow_start=lambda w, i: print(f"Starting: {i}"),
+    on_step_complete=lambda name, r: print(f"{name}: {r.output[:50]}"),
+    on_workflow_complete=lambda w, r: print(f"Done: {r['status']}")
+)
+
+# 6. GUARDRAILS - Validate and retry
+def validate(result):
+    return ("error" not in result.output, "Fix the error")
+
+workflow = Workflow(steps=[
+    WorkflowStep(name="gen", handler=generator, guardrail=validate, max_retries=3)
+])
 ```
 
-### Task Callbacks & Variables (process="workflow")
+### Workflow with Agents
+
+Use Agent objects directly as workflow steps:
 
 ```python
-from praisonaiagents import Agent, Task, PraisonAIAgents
+from praisonaiagents import Agent, Workflow
+from praisonaiagents.workflows import route, parallel
 
-# Per-task agent config (no need to create agent separately)
-task = Task(
-    description="Research {{topic}}",
-    agent_config={"name": "Researcher", "role": "Expert", "llm": "gpt-4o-mini"},
-    variables={"topic": "AI trends"}
+# 1. SEQUENTIAL AGENTS
+researcher = Agent(name="Researcher", role="Research expert", tools=[tavily_search])
+writer = Agent(name="Writer", role="Content writer")
+editor = Agent(name="Editor", role="Editor")
+
+workflow = Workflow(steps=[researcher, writer, editor])
+result = workflow.start("Research and write about AI")
+
+# 2. PARALLEL AGENTS
+workflow = Workflow(steps=[
+    parallel([researcher1, researcher2, researcher3]),
+    aggregator_agent
+])
+
+# 3. ROUTE TO AGENTS
+workflow = Workflow(steps=[
+    classifier_function,
+    route({
+        "technical": [tech_agent],
+        "creative": [creative_agent],
+        "default": [general_agent]
+    })
+])
+
+# 4. WITH PLANNING & REASONING
+workflow = Workflow(
+    steps=[researcher, writer, editor],
+    planning=True,           # Create execution plan
+    planning_llm="gpt-4o",   # LLM for planning
+    reasoning=True,          # Chain-of-thought reasoning
+    verbose=True
 )
 
-# Callbacks for workflow execution
-def on_start(task, task_id):
-    print(f"Starting: {task.name}")
+# 5. TOOLS PER STEP
+workflow = Workflow(steps=[
+    WorkflowStep(
+        name="research",
+        action="Research {{topic}}",
+        tools=[tavily_search, web_scraper],
+        agent_config={"name": "Researcher", "role": "Expert"}
+    )
+])
 
-def on_complete(task, output):
-    print(f"Completed: {task.name}")
+# 6. OUTPUT TO FILE / IMAGES / PYDANTIC
+from pydantic import BaseModel
 
-agents = PraisonAIAgents(
-    agents=[agent],
-    tasks=[task],
-    process="workflow",
-    on_task_start=on_start,
-    on_task_complete=on_complete,
-    variables={"global_var": "shared_value"}  # Global variables for all tasks
+class Report(BaseModel):
+    title: str
+    content: str
+
+workflow = Workflow(steps=[
+    WorkflowStep(name="analyze", action="Analyze image", images=["image.jpg"]),
+    WorkflowStep(name="report", action="Generate report", output_pydantic=Report),
+    WorkflowStep(name="save", action="Save results", output_file="output/report.txt")
+])
+
+# 7. ASYNC EXECUTION
+import asyncio
+
+async def main():
+    result = await workflow.astart("input")
+    print(result)
+
+asyncio.run(main())
+
+# 8. STATUS TRACKING
+workflow.status  # "not_started" | "running" | "completed"
+workflow.step_statuses  # {"step1": "completed", "step2": "skipped"}
+
+# 9. MEMORY CONFIG
+workflow = Workflow(
+    steps=[researcher, writer],
+    memory_config={"provider": "chroma", "persist": True, "collection": "my_workflow"}
 )
+result1 = workflow.start("Research AI")
+result2 = workflow.start("Continue the research")  # Remembers first run
+```
+
+### YAML Workflow Template
+
+```yaml
+# .praison/workflows/research.yaml
+name: Research Workflow
+description: Research and write content with all patterns
+
+agents:
+  researcher:
+    role: Research Expert
+    goal: Find accurate information
+    tools: [tavily_search, web_scraper]
+  writer:
+    role: Content Writer
+    goal: Write engaging content
+  editor:
+    role: Editor
+    goal: Polish content
+
+steps:
+  # Sequential
+  - agent: researcher
+    action: Research {{topic}}
+    output_variable: research_data
+
+  # Routing
+  - name: classifier
+    action: Classify content type
+    route:
+      technical: [tech_handler]
+      creative: [creative_handler]
+      default: [general_handler]
+
+  # Parallel
+  - name: parallel_research
+    parallel:
+      - agent: researcher
+        action: Research market
+      - agent: researcher
+        action: Research competitors
+
+  # Loop
+  - agent: writer
+    action: Write about {{item}}
+    loop_over: topics
+    loop_var: item
+
+  # Repeat (evaluator-optimizer)
+  - agent: editor
+    action: Review and improve
+    repeat:
+      until: "quality > 8"
+      max_iterations: 3
+
+  # Output to file
+  - agent: writer
+    action: Write final report
+    output_file: output/{{topic}}_report.md
+
+variables:
+  topic: AI trends
+  topics: [ML, NLP, Vision]
+
+planning: true
+planning_llm: gpt-4o
+memory_config:
+  provider: chroma
+  persist: true
 ```
 
 ### 9. Hooks
@@ -736,6 +875,12 @@ praisonai workflow run deploy --workflow-var environment=staging --workflow-var
 # Execute with planning mode (AI creates sub-steps for each workflow step)
 praisonai workflow run "Research Blog" --planning --verbose
 
+# Execute with reasoning mode (chain-of-thought)
+praisonai workflow run "Analysis" --reasoning --verbose
+
+# Execute with memory enabled
+praisonai workflow run "Research" --memory
+
 # Show workflow details
 praisonai workflow show deploy
 
@@ -747,8 +892,23 @@ praisonai "What is AI?" --workflow "Research,Summarize" --save
 
 # Inline workflow with step actions
 praisonai "GPT-5" --workflow "Research:Search for info,Write:Write blog" --tools tavily
+
+# Workflow CLI help
+praisonai workflow help
 ```
 
+**Workflow CLI Options:**
+| Flag | Description |
+|------|-------------|
+| `--workflow-var key=value` | Set workflow variable (can be repeated) |
+| `--llm <model>` | LLM model (e.g., openai/gpt-4o-mini) |
+| `--tools <tools>` | Tools (comma-separated, e.g., tavily) |
+| `--planning` | Enable planning mode |
+| `--reasoning` | Enable reasoning mode |
+| `--memory` | Enable memory |
+| `--verbose` | Enable verbose output |
+| `--save` | Save output to file |
+
 ### Hooks CLI:
 ```bash
 # List configured hooks

src/praisonai/praisonai.rb

@@ -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/v2.3.21.tar.gz"
-    sha256 `curl -sL https://github.com/MervinPraison/PraisonAI/archive/refs/tags/v2.3.21.tar.gz | shasum -a 256`.split.first
+    url "https://github.com/MervinPraison/PraisonAI/archive/refs/tags/v2.3.24.tar.gz"
+    sha256 `curl -sL https://github.com/MervinPraison/PraisonAI/archive/refs/tags/v2.3.24.tar.gz | shasum -a 256`.split.first
     license "MIT"
   
     depends_on "[email protected]"

src/praisonai/praisonai/deploy.py

@@ -57,7 +57,7 @@ class CloudDeployer:
             file.write("FROM python:3.11-slim\n")
             file.write("WORKDIR /app\n")
             file.write("COPY . .\n")
-            file.write("RUN pip install flask praisonai==2.3.21 gunicorn markdown\n")
+            file.write("RUN pip install flask praisonai==2.3.24 gunicorn markdown\n")
             file.write("EXPOSE 8080\n")
             file.write('CMD ["gunicorn", "-b", "0.0.0.0:8080", "api:app"]\n')
             

src/praisonai/praisonai/version.py

@@ -1 +1 @@
-__version__ = "2.3.21"
+__version__ = "2.3.24"

src/praisonai/pyproject.toml

@@ -12,7 +12,7 @@ dependencies = [
     "rich>=13.7",
     "markdown>=3.5",
     "pyparsing>=3.0.0",
-    "praisonaiagents>=0.0.199",
+    "praisonaiagents>=0.1.2",
     "python-dotenv>=0.19.0",
     "instructor>=1.3.3",
     "PyYAML>=6.0",

src/praisonai/uv.lock

@@ -4197,7 +4197,7 @@ requires-dist = [
     { name = "praisonai-tools", marker = "extra == 'autogen'", specifier = ">=0.0.22" },
     { name = "praisonai-tools", marker = "extra == 'autogen-v4'", specifier = ">=0.0.22" },
     { name = "praisonai-tools", marker = "extra == 'crewai'", specifier = ">=0.0.22" },
-    { name = "praisonaiagents", specifier = ">=0.0.199" },
+    { name = "praisonaiagents", specifier = ">=0.1.2" },
     { name = "pyautogen", marker = "extra == 'autogen'", specifier = "==0.2.29" },
     { name = "pydantic", marker = "extra == 'chat'", specifier = "<=2.10.1" },
     { name = "pydantic", marker = "extra == 'code'", specifier = "<=2.10.1" },
@@ -4252,7 +4252,7 @@ wheels = [
 
 [[package]]
 name = "praisonaiagents"
-version = "0.0.199"
+version = "0.1.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "aiohttp" },
@@ -4261,9 +4261,9 @@ dependencies = [
     { name = "pydantic" },
     { name = "rich" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/8b/5f/fffbe2473e579edc0bc00022179a896e847128056b6b51905fc497b3cb7d/praisonaiagents-0.0.199.tar.gz", hash = "sha256:276a224350d1a975bdb69f5f07898a60ac9fce0b2d45edbab14e8b755a66853c", size = 538316, upload-time = "2025-12-17T00:01:47.411Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/e3/cd/e00b4dfc4127e2f51acc8c414a392ec45b5e80c45386f082efc31fe7af1b/praisonaiagents-0.1.2.tar.gz", hash = "sha256:e8f94748af81897cd73cee1e74d031296c2103c2c7b8911ee0618c5cfb2a8492", size = 551139, upload-time = "2025-12-17T07:43:27.008Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c9/25/489052a4ed1c8a65a08ae48fabe94b1a399cd82ed2e9c47264a1f82d066c/praisonaiagents-0.0.199-py3-none-any.whl", hash = "sha256:5bc887e097d2e444d63b755d356c537ed30e5f519c2e21adc819160ff4ed5697", size = 515315, upload-time = "2025-12-17T00:01:45.602Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/1f/8a4e5097bee7e8031c2428cf8c1e47b06c85c5f4ce552231de3224d0c301/praisonaiagents-0.1.2-py3-none-any.whl", hash = "sha256:c5a1241da8feb32957b0820ffaa37b03e4ff45462d322fde4ff84f224f10eb91", size = 518797, upload-time = "2025-12-17T07:43:25.172Z" },
 ]
 
 [[package]]