Release v4.7.0

fix: propagate step failures through nested workflow patterns

.github/scripts/merge-gate.js

@@ -74,7 +74,11 @@ function isClaudeTriggerNoise(c) {
 
 function isClaudeFinalReplyComment(c) {
   const login = (c.user?.login || '').toLowerCase();
-  if (!login.includes('praisonai-triage')) return false;
+  const fromAutomation =
+    login.includes('praisonai-triage') ||
+    login.includes('github-actions') ||
+    AUTO_ACTORS.some((a) => a.toLowerCase() === login);
+  if (!fromAutomation) return false;
   return (c.body || '').includes('Claude finished');
 }
 
@@ -96,15 +100,47 @@ function isConflictRebaseTriggerComment(c) {
 
 function isConflictRebaseCompletionComment(c) {
   const login = (c.user?.login || '').toLowerCase();
-  if (!login.includes('praisonai-triage') && !login.includes('github-actions')) return false;
+  const fromAutomation =
+    login.includes('praisonai-triage') ||
+    login.includes('github-actions') ||
+    AUTO_ACTORS.some((a) => a.toLowerCase() === login);
+  if (!fromAutomation) return false;
   const body = (c.body || '').toLowerCase();
   return (
     body.includes('rebase complete') ||
     body.includes('rebase onto') ||
+    body.includes('force-pushed') ||
     (body.includes('conflict') && body.includes('resolved'))
   );
 }
 
+/** HEAD landed after conflict rebase but no FINAL @claude since that push. */
+function needsFinalReReviewAfterConflictRebase(comments, headPushedAt) {
+  if (!headPushedAt) return false;
+  const headTime = new Date(headPushedAt).getTime();
+  const conflictTriggers = comments.filter(isConflictRebaseTriggerComment);
+  if (conflictTriggers.length === 0) return false;
+
+  const latestConflict = conflictTriggers.reduce((a, b) =>
+    new Date(a.created_at) > new Date(b.created_at) ? a : b
+  );
+  const conflictTime = new Date(latestConflict.created_at).getTime();
+  if (headTime <= conflictTime) return false;
+
+  const rebaseDone = comments.some(
+    (c) =>
+      new Date(c.created_at).getTime() > conflictTime && isConflictRebaseCompletionComment(c)
+  );
+  if (!rebaseDone) return false;
+
+  const finalsAfterHead = comments.filter(
+    (c) =>
+      isFinalClaudeTriggerComment(c) &&
+      new Date(c.created_at).getTime() >= headTime - 60000
+  );
+  return finalsAfterHead.length === 0;
+}
+
 function conflictRebaseQuiescent(comments, headPushedAt) {
   const conflictTriggers = comments.filter(isConflictRebaseTriggerComment);
   if (conflictTriggers.length === 0) return true;
@@ -245,6 +281,9 @@ function shouldSkipStaleFinalRecovery(comments, headPushedAt, headPusherLogin =
   if (headPusherLogin && isClaudeAutomationLogin(headPusherLogin)) {
     return { skip: true, reason: 'head pushed by Claude automation' };
   }
+  if (needsFinalReReviewAfterConflictRebase(comments, headPushedAt)) {
+    return { skip: false, reason: '' };
+  }
   if (isPushSoonAfterLatestFinal(comments, headPushedAt)) {
     return {
       skip: true,
@@ -916,6 +955,7 @@ module.exports = {
   hasRecentConflictComment,
   isConflictRebaseTriggerComment,
   isConflictRebaseCompletionComment,
+  needsFinalReReviewAfterConflictRebase,
   conflictRebaseQuiescent,
   hasHumanChangesRequested,
   hasAnyChangesRequested,

.github/workflows/sync-secrets-to-aiui.yml

@@ -24,6 +24,7 @@ jobs:
           CLAUDE_APP_PRIVATE_KEY: ${{ secrets.CLAUDE_APP_PRIVATE_KEY }}
           CLAUDE_CODE_OAUTH_TOKEN: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
           GH_TOKEN_VALUE: ${{ secrets.GH_TOKEN }}
+          DOCS_GITHUB_TOKEN: ${{ secrets.SECRETS_ADMIN_PAT }}
         run: |
           set -euo pipefail
           if [ -z "${GH_TOKEN:-}" ]; then
@@ -34,4 +35,7 @@ jobs:
           echo -n "$CLAUDE_APP_PRIVATE_KEY" | gh secret set CLAUDE_APP_PRIVATE_KEY --repo "$TARGET"
           echo -n "$CLAUDE_CODE_OAUTH_TOKEN" | gh secret set CLAUDE_CODE_OAUTH_TOKEN --repo "$TARGET"
           echo -n "$GH_TOKEN_VALUE" | gh secret set GH_TOKEN --repo "$TARGET"
+          if [ -n "${DOCS_GITHUB_TOKEN:-}" ]; then
+            echo -n "$DOCS_GITHUB_TOKEN" | gh secret set DOCS_GITHUB_TOKEN --repo "$TARGET"
+          fi
           echo "Synced CLAUDE_APP_ID, CLAUDE_APP_PRIVATE_KEY, CLAUDE_CODE_OAUTH_TOKEN, GH_TOKEN → $TARGET"

.github/workflows/test-optimized.yml

@@ -98,6 +98,20 @@ jobs:
           --ignore=tests/unit/cli/test_message_queue.py \
           --maxfail=5
 
+    - name: Run praisonai-agents subscription-auth smoke tests
+      env:
+        PRAISONAI_TEST_TIER: smoke
+        PRAISONAI_ALLOW_NETWORK: '0'
+        OPENAI_MODEL_NAME: gpt-4o-mini
+        OPENAI_API_KEY: 'test-key'
+      shell: bash
+      run: |
+        set -euo pipefail
+        cd src/praisonai-agents
+        python -m pytest tests/smoke/test_subscription_auth.py \
+          tests/unit/auth/test_agent_auth_wiring.py \
+          -q --tb=line --timeout=60 --maxfail=5
+
     - name: Standalone praisonai-code import smoke
       shell: bash
       run: |

README.md

@@ -130,7 +130,7 @@ Each layer wraps the one inside it. When an agent misbehaves, the layer tells yo
 | **3 · Harness** | Can it act, and be checked? | `tools=`, `MCP()`, `guardrails=`, `approval=`, `hooks=`, `sandbox=` |
 | **4 · Loop** | When do we stop? | `execution=ExecutionConfig(...)`, `reflection=`, `autonomy=`, doom-loop detection |
 | **5 · Graph** | Who runs when, and who checks whom? | `AgentFlow`, `route()`, `parallel()`, `loop()`, `repeat()` |
-| **⬡ Managed** | *Where does it actually run?* | `backend=ManagedAgent(compute="e2b")` — remote sandbox, or a fully hosted loop |
+| **⬡ Managed** | *Where does it actually run?* | `tools_run_on="docker"` — one shared sandbox for the tools, or `run_on="anthropic"` for the whole agent |
 
 ### Layer 1 · Prompt — *Did I say it clearly?*
 
@@ -195,7 +195,6 @@ from praisonaiagents import Agent, ExecutionConfig
 agent = Agent(
     instructions="Fix the failing tests.",
     execution=ExecutionConfig(max_iter=30, max_budget=0.50, on_budget_exceeded="stop"),
-    reflection=True,               # completion check — the agent grades its own answer
     autonomy=True,                 # required to drive the loop with run_autonomous()
 )
 result = agent.run_autonomous("Refactor the auth module", max_iterations=5)
@@ -236,22 +235,85 @@ The harness is commoditising; **where** the agent executes is the next multiplie
 pip install praisonai
 ```
 
+The simplest way in is `tools_run_on=` — one whole team or workflow shares **one** sandbox, so a file written by step 1 is there for step 2. Thinking stays on your machine:
+
 ```python
-from praisonai import Agent, ManagedAgent, LocalManagedConfig
+from praisonaiagents import Agent, AgentFlow
 
-# A. Tools run in a remote sandbox; the agent loop stays local
-sandboxed = ManagedAgent(
-    provider="local", compute="e2b",   # or modal | daytona | flyio | docker | tenki
-    config=LocalManagedConfig(model="gpt-4o-mini", name="RemoteTools"),
-)
-agent = Agent(name="builder", backend=sandboxed)
+writer = Agent(name="Writer", instructions="You write files.")
+reader = Agent(name="Reader", instructions="You read files.")
 
-# B. The entire agent loop runs in the cloud (needs ANTHROPIC_API_KEY;
-#    with no key set, ManagedAgent() falls back to a local loop)
-agent = Agent(name="teacher", backend=ManagedAgent())
+flow = AgentFlow(tools_run_on="docker", steps=[writer, reader])  # or e2b | modal | daytona | flyio
+flow.run("Write 'hello' to /workspace/note.txt, then read it back")
+```
+
+Same thing with no Python at all:
+
+```yaml
+name: remote-demo
+tools_run_on: docker      # every step shares one sandbox
+agents:
+  writer: {role: Writer, goal: Write files}
+  reader: {role: Reader, goal: Read files}
+steps:
+  - agent: writer
+    action: "Write 'hello' to /workspace/note.txt"
+  - agent: reader
+    action: "Read /workspace/note.txt"
+```
+
+For a single agent, two words cover it — and they answer different questions:
+
+```python
+from praisonaiagents import Agent
+
+# A. Only the TOOLS move. Thinking stays on your machine.
+agent = Agent(name="builder", instructions="You build things.",
+              tools_run_on="docker")   # docker | e2b | modal | daytona | flyio
+                                       # tenki | sandlock | ssh | novita | subprocess
+
+# B. The WHOLE agent moves — model calls, loop and tools
+agent = Agent(name="teacher", instructions="You teach.", run_on="anthropic")  # hosted
+agent = Agent(name="builder", instructions="You build.", run_on="docker")     # self-hosted
 agent.start("Write a Python script that prints the first 10 primes, then run it")
 ```
 
+Ask any object where it runs, and it will tell you:
+
+```python
+>>> Agent(name="builder", instructions="x", tools_run_on="docker")
+Agent(name='builder', thinks_on='this machine', tools_run_on='a Docker container')
+
+>>> agent.where_does_it_run()
+Thinking (the AI model calls) happens on this machine.
+Tools run on a Docker container.
+Your own tools (check_db) still run on this machine -- only shell, file and
+code tools move. They read and write this machine's files.
+```
+
+Naming a place that cannot do the job is a typo, not a preference, so it says so:
+
+```python
+>>> Agent(name="x", instructions="i", run_on="e2b")
+TypeError: Agent(run_on='e2b') is not valid: run_on= places the whole agent
+-- model calls, loop and tools -- on a managed runtime, and 'e2b' runs
+commands but cannot host an agent loop.
+  To run only the tools there:  Agent(tools_run_on='e2b')
+```
+
+To run one block of code somewhere else, name the place on that call:
+
+```python
+agent.execute_code_sync("print(6 * 7)", run_in="sandlock")   # kernel-enforced
+```
+
+See what is running and reclaim strays:
+
+```bash
+praisonai managed ps          # list running sandboxes
+praisonai managed stop --all  # reclaim them
+```
+
 Sandboxes shut themselves down when idle (`auto_shutdown`, `idle_timeout_s`), and a post-setup snapshot is reused so the next run skips the image pull and dependency install. Commit a `.praisonai/environment.yaml` and the environment travels with the repo.
 
 > 📖 [20 runnable examples](examples/python/managed-agents/) · manage sessions with `praisonai managed sessions list <agent-id>` or `praisonai managed sessions resume <session-id> "<prompt>"`

SECURITY.md

@@ -8,11 +8,18 @@ PraisonAI takes security seriously. We appreciate responsible disclosure from re
 |---------|----------|-------------|
 | `praisonaiagents` | [PyPI](https://pypi.org/project/praisonaiagents/) | `src/praisonai-agents/` |
 | `praisonai` | [PyPI](https://pypi.org/project/praisonai/) | `src/praisonai/` |
-| `praisonai-platform` | [PyPI](https://pypi.org/project/praisonai-platform/) | `src/praisonai-platform/` |
 | `praisonai` | [npm](https://www.npmjs.com/package/praisonai) | `src/praisonai-ts/` |
 
 Report issues against the **specific package** where the vulnerability exists. A platform-only issue does not necessarily affect the core SDK, and vice versa.
 
+### Packages maintained in their own repositories
+
+`praisonai-platform` now lives at
+[MervinPraison/PraisonAI-Platform](https://github.com/MervinPraison/PraisonAI-Platform).
+Report platform vulnerabilities through
+[that repository's advisories](https://github.com/MervinPraison/PraisonAI-Platform/security/advisories/new),
+not here — it is versioned and released independently of the SDK.
+
 ## In scope
 
 - Authentication, authorisation, and multi-tenant isolation (e.g. workspace boundaries)

api.md

@@ -7,7 +7,7 @@ Regenerate with: `praisonai docs api-md --write`
 
 Types:
 ```python
-from praisonaiagents import ContextPolicy, GuardrailResult, HandoffConfig, HandoffCycleError, HandoffDepthError, HandoffError, HandoffInputData, HandoffResult, HandoffTimeoutError, ReflectionOutput, StepResult, TaskOutput, ToolResult, ToolValidationError, WorkflowContext
+from praisonaiagents import ContextPolicy, GuardrailResult, HandoffConfig, HandoffCycleError, HandoffDepthError, HandoffError, HandoffInputData, HandoffResult, HandoffTimeoutError, ReflectionOutput, StepResult, ToolResult, ToolValidationError, WorkflowContext
 ```
 
 Methods:
@@ -16,8 +16,6 @@ Methods:
 * <code title="class HandoffConfig">HandoffConfig.<a href="./src/praisonai-agents/praisonaiagents/agent/handoff.py">from_dict</a>(data: Dict[str, Any]) -> 'HandoffConfig'</code>
 * <code title="class HandoffConfig">HandoffConfig.<a href="./src/praisonai-agents/praisonaiagents/agent/handoff.py">to_dict</a>() -> Dict[str, Any]</code>
 * <code title="class HandoffResult">HandoffResult.<a href="./src/praisonai-agents/praisonaiagents/agent/handoff.py">from_outcome</a>(outcome: AgentRunOutcome, target_agent: Optional[str] = None, source_agent: Optional[str] = None, handoff_depth: int = 0) -> 'HandoffResult'</code>
-* <code title="class TaskOutput">TaskOutput.<a href="./src/praisonai-agents/praisonaiagents/main.py">json</a>() -> Optional[str]</code>
-* <code title="class TaskOutput">TaskOutput.<a href="./src/praisonai-agents/praisonaiagents/main.py">to_dict</a>() -> dict</code>
 * <code title="class ToolResult">ToolResult.<a href="./src/praisonai-agents/praisonaiagents/tools/base.py">is_multimodal</a>() -> bool</code>
 * <code title="class ToolResult">ToolResult.<a href="./src/praisonai-agents/praisonaiagents/tools/base.py">to_dict</a>() -> Dict[str, Any]</code>
 
@@ -32,6 +30,7 @@ Methods:
 
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">aclose</a>() -> None</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">add_mcp_server</a>(name: str, mcp: Any) -> Any</code>
+* <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">add_verification</a>(hook: Any) -> None</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">agent_id</a>() -> str</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">analyze_prompt</a>(prompt: str) -> set</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">auto_memory</a>() -> Optional[bool]</code>
@@ -72,6 +71,7 @@ Methods:
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">redo</a>() -> bool</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">refresh_tools</a>() -> List[Any]</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">remove_mcp_server</a>(name: str) -> bool</code>
+* <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">remove_verification</a>(name: str) -> bool</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">retrieval_config</a>() -> Optional[Any]</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">retrieve</a>(query: str, **kwargs) -> 'ContextPack'</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">rules_manager</a>() -> Optional[Any]</code>
@@ -88,6 +88,7 @@ Methods:
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">thinking_budget</a>(value: Optional[int]) -> None</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">total_cost</a>() -> float</code>
 * <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">undo</a>() -> bool</code>
+* <code title="class Agent">Agent.<a href="./src/praisonai-agents/praisonaiagents/agent/agent.py">where_does_it_run</a>() -> str</code>
 * <code title="class AutoAgents">AutoAgents.<a href="./src/praisonai-agents/praisonaiagents/agents/autoagents.py">astart</a>()</code>
 * <code title="class AutoAgents">AutoAgents.<a href="./src/praisonai-agents/praisonaiagents/agents/autoagents.py">start</a>()</code>
 * <code title="class AutoRagAgent">AutoRagAgent.<a href="./src/praisonai-agents/praisonaiagents/agents/auto_rag_agent.py">achat</a>(message: str, **kwargs) -> str</code>
@@ -464,7 +465,7 @@ Methods:
 
 Types:
 ```python
-from praisonaiagents import SkillLoader, SkillManager, SkillMetadata, SkillProperties, SkillsConfig
+from praisonaiagents import SkillLoader, SkillsConfig
 ```
 
 Methods:
@@ -476,55 +477,6 @@ Methods:
 * <code title="class SkillLoader">SkillLoader.<a href="./src/praisonai-agents/praisonaiagents/skills/loader.py">load_metadata</a>(skill_path: str) -> Optional[LoadedSkill]</code>
 * <code title="class SkillLoader">SkillLoader.<a href="./src/praisonai-agents/praisonaiagents/skills/loader.py">load_references</a>(skill: LoadedSkill) -> dict</code>
 * <code title="class SkillLoader">SkillLoader.<a href="./src/praisonai-agents/praisonaiagents/skills/loader.py">load_scripts</a>(skill: LoadedSkill) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">activate</a>(skill: LoadedSkill) -> bool</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">activate_by_name</a>(name: str) -> bool</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">add_bundle</a>(manifest: 'BundleManifest') -> None</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">add_skill</a>(skill_path: str) -> Optional[LoadedSkill]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">add_skill_by_name</a>(name: str, skill_dirs: Optional[List[str]] = None, include_defaults: bool = True) -> Optional[LoadedSkill]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">approve</a>(identifier: str) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">archive_skill</a>(name: str) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">bundle_names</a>() -> List[str]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">bundles</a>() -> List['BundleManifest']</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">clear</a>() -> None</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">create</a>(name: str, content: str, category: Optional[str] = None, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">create_skill</a>(name: str, content: str, category: str = None, agent_created: bool = True, propose: Optional[bool] = None, base_dir: Optional[str] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">delete</a>(name: str, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">delete_skill</a>(name: str, hard: bool = False, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">discover</a>(skill_dirs: Optional[List[str]] = None, include_defaults: bool = True) -> int</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">discover_bundles</a>(skill_dirs: Optional[List[str]] = None, include_defaults: bool = True) -> int</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">edit</a>(name: str, content: str, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">edit_skill</a>(name: str, content: str, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_allowed_tools</a>(name: str) -> List[str]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_available_skills</a>() -> List[SkillMetadata]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_available_skills_by_state</a>(state: SkillState) -> List[LoadedSkill]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_bundle</a>(name: str) -> Optional['BundleManifest']</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_bundle_instructions</a>() -> List[str]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_instructions</a>(name: str) -> Optional[str]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_skill</a>(name: str) -> Optional[LoadedSkill]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_skills_diagnostics</a>() -> Dict[str, ValidationResult]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">get_user_invocable_skills</a>() -> List[LoadedSkill]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">invoke</a>(name: str, raw_args: str = '', session_id: Optional[str] = None, shell_exec: bool = False) -> Optional[str]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">list_archived_skills</a>() -> List[str]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">list_pending</a>() -> List[dict]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">load_resources</a>(name: str) -> bool</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">patch</a>(name: str, old_string: str, new_string: str, file_path: Optional[str] = None, replace_all: bool = False, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">patch_skill</a>(name: str, old_string: str, new_string: str, file_path: str = None, replace_all: bool = False, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">reject</a>(identifier: str) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">reload</a>() -> Dict[str, List[str]]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">remove_file</a>(name: str, file_path: str, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">remove_skill_file</a>(name: str, file_path: str, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">resolve</a>(selectors: List[str]) -> List[str]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">restore_skill</a>(name: str, skill_dir: Optional[str] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">rollback_skill</a>(name: str) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">skill_names</a>() -> List[str]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">skills</a>() -> List[LoadedSkill]</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">to_prompt</a>() -> str</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">validate_skill_capabilities</a>(skill_name: str, force_refresh: bool = False) -> ValidationResult</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">write_file</a>(name: str, file_path: str, file_content: str, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillManager">SkillManager.<a href="./src/praisonai-agents/praisonaiagents/skills/manager.py">write_skill_file</a>(name: str, file_path: str, file_content: str, propose: Optional[bool] = None) -> dict</code>
-* <code title="class SkillMetadata">SkillMetadata.<a href="./src/praisonai-agents/praisonaiagents/skills/models.py">from_properties</a>(props: SkillProperties) -> 'SkillMetadata'</code>
-* <code title="class SkillProperties">SkillProperties.<a href="./src/praisonai-agents/praisonaiagents/skills/models.py">idle_days</a>() -> Optional[float]</code>
-* <code title="class SkillProperties">SkillProperties.<a href="./src/praisonai-agents/praisonaiagents/skills/models.py">to_dict</a>() -> dict</code>
 * <code title="class SkillsConfig">SkillsConfig.<a href="./src/praisonai-agents/praisonaiagents/config/feature_configs.py">to_dict</a>() -> Dict[str, Any]</code>
 
 # Session
@@ -757,6 +709,7 @@ Methods:
 * <code title="class AgentFlow">AgentFlow.<a href="./src/praisonai-agents/praisonaiagents/workflows/workflows.py">to_dict</a>() -> Dict[str, Any]</code>
 * <code title="class AgentFlow">AgentFlow.<a href="./src/praisonai-agents/praisonaiagents/workflows/workflows.py">verbose</a>() -> bool</code>
 * <code title="class AgentFlow">AgentFlow.<a href="./src/praisonai-agents/praisonaiagents/workflows/workflows.py">verbose</a>(value: bool)</code>
+* <code title="class AgentFlow">AgentFlow.<a href="./src/praisonai-agents/praisonaiagents/workflows/workflows.py">where_does_it_run</a>() -> str</code>
 * <code title="class AgentOSProtocol">AgentOSProtocol.<a href="./src/praisonai-agents/praisonaiagents/app/protocols.py">get_app</a>() -> Any</code>
 * <code title="class AgentOSProtocol">AgentOSProtocol.<a href="./src/praisonai-agents/praisonaiagents/app/protocols.py">serve</a>(host: Optional[str] = None, port: Optional[int] = None, reload: bool = False, **kwargs: Any) -> None</code>
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">aannounce_completion</a>(agent_id: str, task_id: str, result: Any, success: bool = True, error: Optional[str] = None, metadata: Optional[Dict[str, Any]] = None) -> None</code>
@@ -793,7 +746,7 @@ Methods:
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">get_token_usage_summary</a>() -> Dict[str, Any]</code>
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">has_state</a>(key: str) -> bool</code>
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">increment_state</a>(key: str, amount: float = 1, default: float = 0) -> float</code>
-* <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">launch</a>(path: str = '/agents', port: int = 8000, host: str = '0.0.0.0', debug: bool = False, protocol: str = 'http')</code>
+* <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">launch</a>(path: str = '/agents', port: int = 8000, host: str = '127.0.0.1', debug: bool = False, protocol: str = 'http')</code>
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">restore_session_state</a>(session_id: str) -> bool</code>
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">run</a>(content = None, return_dict = False, **kwargs)</code>
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">run_all_tasks</a>()</code>
@@ -809,6 +762,7 @@ Methods:
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">update_plan_step_status</a>(step_id: str, status: str) -> bool</code>
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">update_state</a>(updates: Dict) -> None</code>
 * <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">wait_for_completions</a>(timeout: Optional[float] = None, agent_ids: Optional[List[str]] = None) -> List[SubAgentCompletionEvent]</code>
+* <code title="class AgentTeam">AgentTeam.<a href="./src/praisonai-agents/praisonaiagents/agents/agents.py">where_does_it_run</a>() -> str</code>
 * <code title="class AutoApproveBackend">AutoApproveBackend.<a href="./src/praisonai-agents/praisonaiagents/approval/backends.py">request_approval</a>(request: ApprovalRequest) -> ApprovalDecision</code>
 * <code title="class AutoApproveBackend">AutoApproveBackend.<a href="./src/praisonai-agents/praisonaiagents/approval/backends.py">request_approval_sync</a>(request: ApprovalRequest) -> ApprovalDecision</code>
 * <code title="class RunOutcome">RunOutcome.<a href="./src/praisonai-agents/praisonaiagents/agent/run_outcome.py">completed</a>(output: Optional[str] = None) -> 'RunOutcome'</code>
@@ -890,11 +844,13 @@ Methods:
 * <code title="cli">praisonai managed list <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai managed managed-callback <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai managed multi <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
+* <code title="cli">praisonai managed ps <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai managed restore <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai managed resume <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai managed run <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai managed save <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai managed show <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
+* <code title="cli">praisonai managed stop <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai managed update <a href="./src/praisonai/praisonai/cli/commands/managed.py">--help</a></code>
 * <code title="cli">praisonai n8n export <a href="./src/praisonai/praisonai/cli/commands/n8n.py">--help</a></code>
 * <code title="cli">praisonai n8n import <a href="./src/praisonai/praisonai/cli/commands/n8n.py">--help</a></code>
@@ -924,6 +880,7 @@ Methods:
 * <code title="cli">praisonai recipe list <a href="./src/praisonai/praisonai/cli/commands/recipe.py">--help</a></code>
 * <code title="cli">praisonai recipe optimize <a href="./src/praisonai/praisonai/cli/commands/recipe.py">--help</a></code>
 * <code title="cli">praisonai recipe run <a href="./src/praisonai/praisonai/cli/commands/recipe.py">--help</a></code>
+* <code title="cli">praisonai recipe runs <a href="./src/praisonai/praisonai/cli/commands/recipe.py">--help</a></code>
 * <code title="cli">praisonai recipe serve <a href="./src/praisonai/praisonai/cli/commands/recipe.py">--help</a></code>
 * <code title="cli">praisonai replay cleanup <a href="./src/praisonai/praisonai/cli/commands/replay.py">--help</a></code>
 * <code title="cli">praisonai replay context <a href="./src/praisonai/praisonai/cli/commands/replay.py">--help</a></code>
@@ -940,7 +897,11 @@ Methods:
 * <code title="cli">praisonai schedule describe <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai schedule list <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai schedule logs <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
+* <code title="cli">praisonai schedule pause <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
+* <code title="cli">praisonai schedule remove <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai schedule restart <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
+* <code title="cli">praisonai schedule resume <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
+* <code title="cli">praisonai schedule runs <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai schedule schedule-callback <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai schedule start <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai schedule stats <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
@@ -949,6 +910,7 @@ Methods:
 * <code title="cli">praisonai schedule suggestion-dismiss <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai schedule suggestion-propose <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai schedule suggestions <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
+* <code title="cli">praisonai schedule update <a href="./src/praisonai/praisonai/cli/commands/schedule.py">--help</a></code>
 * <code title="cli">praisonai validate check <a href="./src/praisonai/praisonai/cli/commands/validate.py">--help</a></code>
 * <code title="cli">praisonai validate schema <a href="./src/praisonai/praisonai/cli/commands/validate.py">--help</a></code>
 * <code title="cli">praisonai validate validate <a href="./src/praisonai/praisonai/cli/commands/validate.py">--help</a></code>

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>=4.6.162" \
+    "praisonai>=4.7.0" \
     "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>=4.6.162" \
+    "praisonai>=4.7.0" \
     "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>=4.6.162" \
+    "praisonai>=4.7.0" \
     "praisonai[ui]" \
     "praisonai[crewai]"
 

examples/persistence/state_redis.py

@@ -14,7 +14,7 @@ Expected Output:
     Agent responds with state persisted to Redis
 """
 
-from praisonaiagents import Agent
+from praisonaiagents import Agent, MemoryConfig
 from praisonaiagents.db import db
 
 print("=== Redis State Store (Agent-First) ===")
@@ -28,8 +28,7 @@ my_db = db(
 agent = Agent(
     name="Assistant",
     instructions="You are a helpful assistant.",
-    memory=my_db,
-    session_id="redis-state-example"
+    memory=MemoryConfig(db=my_db, session_id="redis-state-example"),
 )
 
 # Chat - state is automatically managed via Redis

examples/python/autonomy/03_verification_hooks.py

@@ -23,6 +23,7 @@ from praisonaiagents.hooks.verification import (
     VerificationResult,
     BaseVerificationHook,
     CommandVerificationHook,
+    FileCheckHook,
 )
 
 
@@ -100,7 +101,19 @@ def main():
     print(f"   Name: {cmd_hook.name}")
     print(f"   Command: {cmd_hook.command}")
     print(f"   Timeout: {cmd_hook.timeout_seconds}s")
-    
+
+    # 6. FileCheckHook: declarative, no-shell completion gate
+    print("\n6. FileCheckHook (declarative completion gate):")
+    file_hook = FileCheckHook(
+        name="changelog", path="CHANGELOG.md", non_empty=True
+    )
+    file_result = file_hook.run()
+    print(f"   {file_hook.name}: success={file_result.success} — {file_result.output}")
+    print(
+        "   Gate wiring: when verification_hooks are configured, the autonomous "
+        "loop cannot finalise with a success outcome while a blocking hook fails."
+    )
+
     print("\n" + "=" * 60)
     print("✓ Verification hooks example completed successfully!")
     print("=" * 60)

examples/python/mcp/filesystem-mcp.py

@@ -7,12 +7,12 @@ allowed_dirs = [
     "/path/to/other/allowed/dir"
 ]
 
-# Use a single string command with allowed directories
+# Pass the executable as the first token and the allowed directories as args
 filesystem_agent = Agent(
     instructions="""You are a helpful assistant that can interact with the filesystem.
     Use the available tools when relevant to manage files and directories.""",
     llm="gpt-4o-mini",
-    tools=MCP("npx -y @modelcontextprotocol/server-filesystem", args=allowed_dirs)
+    tools=MCP("npx", args=["-y", "@modelcontextprotocol/server-filesystem", *allowed_dirs])
 )
 
 filesystem_agent.start("List files in the allowed directories") 

examples/serve/README.md

@@ -24,7 +24,7 @@ Examples for all PraisonAI server types and the unified endpoints CLI.
 | `praisonai serve unified --port 8765` | Start unified server |
 | `praisonai serve agents --file agents.yaml --port 8000` | Start agents API |
 | `praisonai serve recipe --port 8765` | Start recipe server |
-| `praisonai serve mcp --transport http --port 8080` | Start MCP server |
+| `praisonai mcp serve --transport http-stream --port 8080` | Start MCP server |
 | `praisonai serve tools --port 8081` | Start tools MCP server |
 | `praisonai serve a2a --port 8082` | Start A2A server |
 | `praisonai serve a2u --port 8083` | Start A2U server |

examples/serve/mcp_http_server.py

@@ -13,7 +13,7 @@ Usage:
     python mcp_http_server.py
     
     # Or use CLI:
-    praisonai serve mcp --transport http --port 8080
+    praisonai mcp serve --transport http-stream --port 8080
     
     # Test with curl:
     curl http://localhost:8080/mcp/tools

examples/terminal_bench/job_code_smoke.yaml

@@ -14,9 +14,9 @@ datasets:
     # A small verified subset for a cheap smoke run. Verify against the registry
     # with `harbor datasets list` before editing.
     task_names:
-      - hello-world
-      - fix-permissions
-      - csv-to-parquet
+      - terminal-bench/build-pmars
+      - terminal-bench/bn-fit-modify
+      - terminal-bench/break-filter-js-from-html
 
 agents:
   - import_path: examples.terminal_bench.praisonai_code_agent:PraisonAICodeAgent

examples/terminal_bench/praisonai_code_agent.py

@@ -81,17 +81,19 @@ class PraisonAICodeAgent(BaseInstalledAgent):
         model = self.model_name or "openai/gpt-4o-mini"
 
         command = (
+            "log_file=$(mktemp /tmp/praisonai_code.XXXXXX); "
+            "chmod 600 \"$log_file\"; "
             f"praisonai code -p --output json {shlex.quote(instruction)} "
             f"--dangerously-skip-approval "
             f"--model {shlex.quote(model)} "
-            "> /tmp/praisonai_code.log 2>&1; "
+            "> \"$log_file\" 2>&1; "
             # Capture the real exit status so install/auth/startup failures still
             # surface instead of being silently masked (Harbor otherwise records a
             # completed attempt for a crashed agent).
             "status=$?; "
             # Echo the envelope so Harbor captures it for post-run accounting.
-            "cp /tmp/praisonai_code.log /tmp/praisonai_code_run.log 2>/dev/null || true; "
-            "cat /tmp/praisonai_code.log; "
+            "cat \"$log_file\"; "
+            "rm -f \"$log_file\"; "
             # `praisonai code` normally exits 0 (Harbor grades by task
             # verification), so a genuine benchmark miss won't fail here — but a
             # nonzero status means the assistant itself failed to run.
@@ -128,7 +130,6 @@ class PraisonAICodeAgent(BaseInstalledAgent):
             "agent_type": "code-cli",
             "agent_name": self.name(),
             "model": self.model_name,
-            "log_path": "/tmp/praisonai_code.log",
             "session_id": envelope.get("session_id") if envelope else None,
             "status": envelope.get("status") if envelope else "unknown",
         }

examples/terminal_bench/praisonai_external_agent.py

@@ -34,6 +34,7 @@ except ImportError as e:
 try:
     from praisonaiagents import Agent
     from praisonaiagents.approval import get_approval_registry, AutoApproveBackend
+    from praisonaiagents.config.feature_configs import ExecutionConfig
 except ImportError as e:
     raise ImportError(
         f"PraisonAI not installed: {e}\n"
@@ -41,6 +42,13 @@ except ImportError as e:
     ) from e
 
 
+TB_EXECUTION = ExecutionConfig(
+    max_tool_calls_per_turn=80,
+    max_iter=40,
+    max_steps=80,
+)
+
+
 class PraisonAIExternalAgent(BaseAgent):
     """External PraisonAI agent that drives a Harbor container environment."""
 
@@ -142,26 +150,14 @@ class PraisonAIExternalAgent(BaseAgent):
                 ),
                 tools=[bash_tool],
                 llm=self.model_name or "openai/gpt-4o-mini",
+                execution=TB_EXECUTION,
             )
 
-            # Execute the agent with outer loop to handle premature stopping
+            # Execute one deliberately long coding turn. Harbor's hidden verifier
+            # is the only completion authority; model or pytest text is not.
             print(f"🚀 PraisonAI Agent starting task: {instruction[:100]}...")
             result = await agent.achat(instruction)
-            for _iter in range(19):
-                result_str = str(result)
-                # Stop if test passed
-                if any(sig in result_str.lower() for sig in [
-                    " passed", "passed ", "test passed", "all tests pass", "1 passed"
-                ]):
-                    break
-                result = await agent.achat(
-                    "The task is NOT complete yet — keep working. Run bash_tool commands now. "
-                    "If you haven't already: (1) read all files in /app/, "
-                    "(2) run the test to see the exact failure, "
-                    "(3) implement a fix, (4) run the test again. "
-                    "Repeat until the test passes. What is your next bash_tool command?"
-                )
-            print(f"✅ PraisonAI Agent completed task")
+            print("✅ PraisonAI Agent completed task")
             
             # Populate Harbor context with metadata
             self._populate_context(agent, context, result)
@@ -203,6 +199,10 @@ class PraisonAIExternalAgent(BaseAgent):
                 "tools_used": ["bash_tool"],
                 "framework": "praisonai",
                 "version": self.version(),
+                "stop_reason": getattr(agent, "last_stop_reason", None),
+                "max_tool_calls_per_turn": TB_EXECUTION.max_tool_calls_per_turn,
+                "max_iter": TB_EXECUTION.max_iter,
+                "max_steps": TB_EXECUTION.max_steps,
             }
             
         except Exception as e:
@@ -218,4 +218,4 @@ if __name__ == "__main__":
     print("  --model openai/gpt-4o-mini")
     print()
     print("Dependencies:")
-    print("  pip install harbor praisonaiagents")
+    print("  pip install harbor praisonaiagents")

examples/terminal_bench/test_adapter_config.py

@@ -0,0 +1,39 @@
+"""Static regression checks for the in-tree Harbor adapter configuration."""
+
+from pathlib import Path
+
+import yaml
+
+
+EXAMPLES = Path(__file__).parent
+
+
+def test_external_adapter_sets_coding_execution_budget():
+    source = (EXAMPLES / "praisonai_external_agent.py").read_text(encoding="utf-8")
+
+    assert "ExecutionConfig" in source
+    assert "max_tool_calls_per_turn=80" in source
+    assert "max_iter=40" in source
+    assert "max_steps=80" in source
+
+
+def test_external_adapter_does_not_guess_completion_from_pytest_text():
+    source = (EXAMPLES / "praisonai_external_agent.py").read_text(encoding="utf-8")
+
+    assert '"1 passed"' not in source
+    assert '"test passed"' not in source
+
+
+def test_smoke_yaml_uses_terminal_bench_21_registry_ids():
+    job = yaml.safe_load(
+        (EXAMPLES / "job_code_smoke.yaml").read_text(encoding="utf-8")
+    )
+    names = job["datasets"][0]["task_names"]
+
+    assert names
+    assert all(name.startswith("terminal-bench/") for name in names)
+    assert names == [
+        "terminal-bench/build-pmars",
+        "terminal-bench/bn-fit-modify",
+        "terminal-bench/break-filter-js-from-html",
+    ]

examples/yaml/workflows/remote_sandbox_workflow.yaml

@@ -0,0 +1,45 @@
+# Run a whole workflow's shell/file tools inside ONE shared remote sandbox.
+#
+#   praisonai examples/yaml/workflows/remote_sandbox_workflow.yaml
+#
+# `tools_run_on:` is the only line that differs from a normal local workflow.
+# Every step shares the same sandbox and working directory, so the file the
+# writer creates is there for the reader. Thinking and orchestration still run
+# on your machine, and the sandbox is torn down when the run finishes.
+#
+# Places: docker | e2b | modal | daytona | flyio | tenki | local
+#         subprocess | sandlock | ssh | novita
+#   - docker needs a local Docker daemon and nothing else
+#   - e2b / modal / daytona / flyio need that provider's API key
+#   - sandlock is a kernel-enforced local boundary (Linux 6.12+)
+#   - subprocess is process separation only, NOT a security boundary
+#
+# `run_on:` is still accepted here as the old spelling for this key.
+#
+# See what is still running:  praisonai managed ps
+# Reclaim strays:             praisonai managed stop --all
+
+name: remote-sandbox-demo
+tools_run_on: docker
+
+agents:
+  writer:
+    role: File Writer
+    goal: Write files into the shared workspace
+    instructions: >
+      You have write_file and execute_command running on a remote Linux
+      sandbox. Use them; never say you cannot run commands.
+
+  reader:
+    role: File Reader
+    goal: Read files back out of the shared workspace
+    instructions: >
+      You have read_file and execute_command running on the same remote
+      sandbox the writer used. Use them.
+
+steps:
+  - agent: writer
+    action: "Use write_file to put exactly 'hello from step 1' in /workspace/note.txt"
+
+  - agent: reader
+    action: "Use read_file on /workspace/note.txt and report its exact contents"

src/praisonai-agents/.agent/workflows/security-audit.md

@@ -99,13 +99,12 @@ pip index versions praisonaiagents | head -1
 pip index versions praisonai | head -1
 ```
 
-13. Publish **praisonai-platform** (only when the advisory targets platform code):
-```bash
-cd /Users/praison/praisonai-package/src/praisonai-platform
-praisonai publish pypi
-# or: uv lock && uv build && uv publish --token $PYPI_TOKEN
-```
-Platform releases are **independent** of `praisonai` / `praisonaiagents` version numbers. A platform-only fix (e.g. `0.1.8`) does **not** require bumping the main SDK.
+13. **praisonai-platform** advisories are handled in a different repository:
+[MervinPraison/PraisonAI-Platform](https://github.com/MervinPraison/PraisonAI-Platform).
+It is not part of this monorepo and is not published from here. Triage, patch,
+release and advisory publication all happen in that repo, on its own semver line.
+Nothing in this repository imports it, so a platform advisory never requires an
+SDK release.
 
 14. Publish **praisonai (npm / TypeScript)** (only when the advisory targets `src/praisonai-ts`):
 ```bash
@@ -113,9 +112,8 @@ cd /Users/praison/praisonai-package/src/praisonai-ts
 npm version patch && npm publish
 ```
 
-15. Verify platform / npm when applicable:
+15. Verify npm when applicable:
 ```bash
-pip index versions praisonai-platform | head -1
 npm view praisonai version
 ```
 
@@ -139,7 +137,7 @@ gh api repos/MervinPraison/PraisonAI/security-advisories/<GHSA_ID> \
 }
 EOF
 ```
-- `<PACKAGE_NAME>`: `praisonaiagents` (core SDK), `praisonai` (Python wrapper), `praisonai-platform` (platform layer), or `praisonai` (npm ecosystem for TypeScript)
+- `<PACKAGE_NAME>`: `praisonaiagents` (core SDK), `praisonai` (Python wrapper), or `praisonai` (npm ecosystem for TypeScript)
 - `<LAST_VULNERABLE_VERSION>`: version before the fix
 - `<FIRST_PATCHED_VERSION>`: the version just published with the fix
 - Always include credits for the reporter
@@ -190,8 +188,8 @@ After validating each advisory in source code, assign **one** label:
 | Package | Version line | Notes |
 |---|---|---|
 | `praisonaiagents` + `praisonai` | Aligned patch (e.g. `1.6.62` / `4.6.62`) | Main SDK + wrapper; bump together via `bump_and_release.py` |
-| `praisonai-platform` | Independent semver (e.g. `0.1.8`) | Platform-only fixes; users need **not** upgrade the main SDK |
 | `praisonai` (npm) | Independent semver (e.g. `1.7.2`) | TypeScript SDK; publish from `src/praisonai-ts` |
+| `praisonai-platform` | Independent semver, **separate repository** | Not released from this repo — see [PraisonAI-Platform](https://github.com/MervinPraison/PraisonAI-Platform) |
 
 Always set the advisory `patched_versions` to the **package that actually contains the fix**. Do not imply users must upgrade all packages when only one changed.
 
@@ -217,9 +215,11 @@ Close (or publish advisory with no code change only if already mitigated) when:
 |---|---|---|---|
 | praisonaiagents | pip | src/praisonai-agents/praisonaiagents/ | tools/, mcp/, memory/, agent/, sandbox/ |
 | praisonai | pip | src/praisonai/praisonai/ | cli/, capabilities/, mcp_server/, sandbox/, recipe/ |
-| praisonai-platform | pip | src/praisonai-platform/praisonai_platform/ | api/routes/, services/, auth |
 | praisonai | npm | src/praisonai-ts/src/ | agent/, mcp/, tools/, code-mode/ |
 
+`praisonai-platform` is **not audited from this repository** — it lives in
+[MervinPraison/PraisonAI-Platform](https://github.com/MervinPraison/PraisonAI-Platform).
+
 ## RESOURCES & DOCUMENTATION
 
 - **Official Repository**: https://github.com/MervinPraison/PraisonAI — Source code, issue tracking, and contribution guidelines
@@ -234,7 +234,7 @@ Close (or publish advisory with no code change only if already mitigated) when:
 - ALWAYS set patched_versions before publishing advisory
 - ALWAYS publish to PyPI/npm BEFORE publishing advisories (so Dependabot can point users to a fix)
 - Python SDK: `praisonai publish pypi` from `src/praisonai-agents`, then `bump_and_release.py` for the wrapper
-- Platform-only: `praisonai publish pypi` from `src/praisonai-platform` — no SDK bump required
+- Platform-only: released from the separate [PraisonAI-Platform](https://github.com/MervinPraison/PraisonAI-Platform) repo, never from here
 - npm: `npm publish` from `src/praisonai-ts` when the TypeScript SDK is affected
 - ALWAYS request CVE AFTER publishing advisory
 - Use generic commit messages: "refactor: harden X" not "fix: SQL injection in Y"

src/praisonai-agents/AGENTS.md

@@ -1074,7 +1074,7 @@ PraisonAI's reliability engineering rests on three pillars — **C**ontext, **H*
 > **Evaluator status.** The `ContextEvaluator` / `HarnessEvaluator` / `LoopEvaluator` classes named in the rubric are **planned follow-ups** (PA-CHL-001–004), not yet in the repo. Until they land, the equivalent checks are covered by existing modules:
 >
 > - **Context** — `context/tokens.py` and `eval/tokens.py` (`estimate_tokens` / `count_tokens`) for budget; `compaction/compactor.py` (`ContextCompactor`) for compaction/retention.
-> - **Harness** — `runtime/turn_context.py` for turn-context parity; `runtime/example_harness.py` and `tests/` for harness/artifact checks.
+> - **Harness** — `runtime/turn_context.py` for turn-context parity; `examples/runtime/example_harness.py` and `tests/` for harness/artifact checks.
 > - **Loop** — `eval/loop.py` (`EvaluationLoop`) and `agent/autonomy.py` for convergence and doom-loop guards.
 >
 > Keep terminology consistent with the `agent/context_agent.py` (`ContextAgent`) PRP methodology.

src/praisonai-agents/DOCS_PARITY.md

@@ -18,76 +18,76 @@ This report compares **Python SDK feature categories** against **Python document
 | Category | Features | Docs | Lines |
 |----------|----------|------|-------|
 | ✅ AGUI | 1 | 2 | 420 |
-| ✅ Agent | 22 | 50 | 15809 |
+| ✅ Agent | 22 | 53 | 16965 |
 | ✅ Agent-to-Agent (A2A) | 1 | 7 | 2996 |
 | ✅ Agent-to-User (A2U) | 1 | 3 | 689 |
-| ✅ Approval | 1 | 5 | 2537 |
+| ✅ Approval | 1 | 5 | 2658 |
 | ✅ Audio | 2 | 12 | 1000 |
-| ✅ Auto Generation | 5 | 10 | 3049 |
-| ✅ Autonomy | 3 | 7 | 2744 |
-| ✅ Bots | 7 | 31 | 13314 |
+| ✅ Auto Generation | 5 | 11 | 3542 |
+| ✅ Autonomy | 3 | 7 | 2763 |
+| ✅ Bots | 7 | 31 | 13403 |
 | ✅ Budget | 1 | 1 | 287 |
-| ✅ CLI | 5 | 121 | 36021 |
-| ✅ Chunking | 2 | 2 | 385 |
+| ✅ CLI | 5 | 122 | 36654 |
+| ✅ Chunking | 2 | 2 | 420 |
 | ✅ Citations | 2 | 1 | 202 |
-| ✅ Code Execution | 2 | 14 | 4583 |
-| ✅ Conditions | 1 | 3 | 1179 |
-| ✅ Configuration | 3 | 8 | 3047 |
-| ✅ Context Management | 16 | 34 | 13396 |
+| ✅ Code Execution | 2 | 14 | 4851 |
+| ✅ Conditions | 1 | 3 | 1321 |
+| ✅ Configuration | 3 | 8 | 3140 |
+| ✅ Context Management | 16 | 36 | 13890 |
 | ✅ Deep Research | 8 | 2 | 587 |
 | ✅ Display | 6 | 3 | 741 |
 | ✅ Embeddings | 6 | 23 | 2049 |
 | ✅ Evaluation | 1 | 7 | 2905 |
 | ✅ Events | 1 | 2 | 749 |
-| ✅ Execution | 3 | 4 | 1562 |
+| ✅ Execution | 3 | 4 | 1586 |
 | ✅ Failover | 2 | 1 | 432 |
-| ✅ Files | 2 | 7 | 2273 |
+| ✅ Files | 2 | 7 | 2275 |
 | ✅ Flow | 1 | 3 | 788 |
-| ✅ Gateway | 7 | 72 | 26391 |
-| ✅ Guardrails | 4 | 4 | 2017 |
-| ✅ Handoffs | 11 | 6 | 2563 |
-| ✅ Hooks | 2 | 9 | 4317 |
+| ✅ Gateway | 7 | 78 | 28577 |
+| ✅ Guardrails | 4 | 4 | 2205 |
+| ✅ Handoffs | 11 | 6 | 2567 |
+| ✅ Hooks | 2 | 9 | 4488 |
 | ✅ Image | 1 | 11 | 1255 |
-| ✅ Knowledge | 4 | 13 | 4333 |
-| ✅ LLM | 3 | 15 | 5192 |
-| ✅ Loops | 4 | 5 | 1422 |
-| ✅ MCP | 1 | 61 | 13847 |
-| ✅ Memory | 6 | 18 | 7070 |
+| ✅ Knowledge | 4 | 16 | 4942 |
+| ✅ LLM | 3 | 15 | 5296 |
+| ✅ Loops | 4 | 5 | 1475 |
+| ✅ MCP | 1 | 61 | 14135 |
+| ✅ Memory | 6 | 18 | 7133 |
 | ✅ OCR | 2 | 1 | 237 |
 | ✅ Observability | 2 | 23 | 2814 |
 | ✅ Optimizer | 1 | 2 | 792 |
-| ✅ Output | 3 | 5 | 1201 |
+| ✅ Output | 3 | 5 | 1246 |
 | ✅ Parallel Execution | 3 | 2 | 497 |
-| ✅ Planning | 6 | 6 | 1549 |
-| ✅ Plugins | 8 | 4 | 2850 |
-| ✅ Prompts | 2 | 9 | 1965 |
-| ✅ Providers | 1 | 56 | 7735 |
+| ✅ Planning | 6 | 6 | 1605 |
+| ✅ Plugins | 8 | 5 | 3120 |
+| ✅ Prompts | 2 | 9 | 1971 |
+| ✅ Providers | 1 | 56 | 7781 |
 | ✅ Query | 1 | 2 | 736 |
-| ✅ RAG | 5 | 15 | 3003 |
+| ✅ RAG | 5 | 15 | 3118 |
 | ✅ Realtime | 2 | 5 | 668 |
-| ✅ Reflection | 3 | 3 | 680 |
+| ✅ Reflection | 3 | 3 | 683 |
 | ✅ Retrieval | 2 | 5 | 1054 |
 | ✅ Routing | 1 | 2 | 430 |
-| ✅ Sandbox | 5 | 6 | 2918 |
-| ✅ Security | 1 | 3 | 3206 |
-| ✅ Sessions | 4 | 17 | 6963 |
+| ✅ Sandbox | 5 | 7 | 3483 |
+| ✅ Security | 1 | 3 | 3415 |
+| ✅ Sessions | 4 | 19 | 7576 |
 | ✅ Skills | 6 | 15 | 5638 |
 | ✅ Tasks | 2 | 6 | 2819 |
 | ✅ Telemetry | 1 | 2 | 604 |
 | ✅ Templates | 1 | 8 | 1663 |
-| ✅ Tools | 12 | 141 | 41242 |
+| ✅ Tools | 12 | 144 | 42732 |
 | ✅ Tracing | 3 | 2 | 139 |
-| ✅ Vector Store | 1 | 12 | 1211 |
+| ✅ Vector Store | 1 | 12 | 1215 |
 | ✅ Video | 2 | 6 | 649 |
 | ✅ Vision | 2 | 1 | 329 |
-| ✅ Web | 3 | 9 | 2467 |
-| ✅ Workflows | 5 | 16 | 6589 |
+| ✅ Web | 3 | 9 | 2484 |
+| ✅ Workflows | 5 | 17 | 7049 |
 
 ## Documentation Without Features
 
 These docs exist but don't match any implemented feature category:
 
-- ℹ️ Database (42 docs, 6713 lines)
+- ℹ️ Database (42 docs, 6766 lines)
 - ℹ️ Documents (1 docs, 787 lines)
 
 ---

src/praisonai-agents/examples/runtime/example_harness.py

@@ -14,16 +14,16 @@ import logging
 import time
 from typing import Any, Dict, List, Optional, TYPE_CHECKING
 
-from .turn_context import (
+from praisonaiagents.runtime.turn_context import (
     PreparedTurnContext,
-    TurnRuntimeProtocol,
     RuntimeMode,
 )
-from .context_builder import default_context_builder
-from ..streaming.events import StreamEvent, StreamEventType
+from praisonaiagents.runtime.protocols import TurnRuntimeProtocol
+from praisonaiagents.runtime.context_builder import default_context_builder
+from praisonaiagents.streaming.events import StreamEvent, StreamEventType
 
 if TYPE_CHECKING:
-    from ..agent.protocols import AgentProtocol
+    from praisonaiagents.agent.protocols import AgentProtocol
 
 logger = logging.getLogger(__name__)
 
@@ -363,4 +363,4 @@ async def demonstrate_harness_integration():
 
 if __name__ == "__main__":
     # Run the demonstration
-    asyncio.run(demonstrate_harness_integration())
+    asyncio.run(demonstrate_harness_integration())

src/praisonai-agents/praisonaiagents/agent/async_safety.py

@@ -110,14 +110,32 @@ class AsyncSafeState:
     def __init__(self, initial_value: Any = None):
         self.value = initial_value
         self._lock = DualLock()
+        self._activity_lock = threading.Lock()
+        self._async_holders = 0
+        self._copying = False
 
     def __deepcopy__(self, memo):
-        """Copy the protected value while replacing all lock state."""
+        """Copy protected state, rejecting overlap with asynchronous access."""
         result = type(self).__new__(type(self))
         memo[id(self)] = result
-        with self.lock():
-            result.value = copy.deepcopy(self.value, memo)
+        with self._activity_lock:
+            if self._async_holders:
+                raise RuntimeError(
+                    "Cannot deepcopy AsyncSafeState during asynchronous access"
+                )
+            if self._copying:
+                raise RuntimeError("AsyncSafeState deepcopy is already in progress")
+            self._copying = True
+        try:
+            with self.lock():
+                result.value = copy.deepcopy(self.value, memo)
+        finally:
+            with self._activity_lock:
+                self._copying = False
         result._lock = DualLock()
+        result._activity_lock = threading.Lock()
+        result._async_holders = 0
+        result._copying = False
         return result
         
     @contextmanager 
@@ -130,7 +148,25 @@ class AsyncSafeState:
     async def async_lock(self):
         """Acquire lock in async context."""
         async with self._lock.async_lock():
-            yield self.value
+            self._begin_async_access()
+            try:
+                yield self.value
+            finally:
+                self._end_async_access()
+
+    def _begin_async_access(self) -> None:
+        """Register async access unless a synchronous copy is in progress."""
+        with self._activity_lock:
+            if self._copying:
+                raise RuntimeError(
+                    "Cannot access AsyncSafeState during synchronous deepcopy"
+                )
+            self._async_holders += 1
+
+    def _end_async_access(self) -> None:
+        """Unregister one active asynchronous accessor."""
+        with self._activity_lock:
+            self._async_holders -= 1
             
     def __enter__(self):
         """Support for synchronous context manager protocol (backward compatibility)."""
@@ -146,12 +182,20 @@ class AsyncSafeState:
         """Support for asynchronous context manager protocol."""
         async_lock = self._lock._get_async_lock()
         await async_lock.acquire()
+        try:
+            self._begin_async_access()
+        except Exception:
+            async_lock.release()
+            raise
         return self.value
         
     async def __aexit__(self, exc_type, exc_val, exc_tb):
         """Support for asynchronous context manager protocol."""
         async_lock = self._lock._get_async_lock()
-        async_lock.release()
+        try:
+            self._end_async_access()
+        finally:
+            async_lock.release()
         return None
             
     def get(self) -> Any:

src/praisonai-agents/praisonaiagents/agent/audio_agent.py

@@ -136,9 +136,8 @@ class AudioAgent:
                 - AudioConfig: Full configuration object
             verbose: Verbosity level for output
         """
-        # Handle model= alias
-        if llm is None and model is not None:
-            llm = model
+        from ..utils.model_alias import resolve_model_name
+        llm = resolve_model_name(llm, model, "AudioAgent")
         
         self.name = name or "AudioAgent"
         self.instructions = instructions

src/praisonai-agents/praisonaiagents/agent/autonomy.py

@@ -106,8 +106,11 @@ class AutonomyConfig:
         track_changes: Whether to track filesystem changes via shadow git.
             Defaults to True when level="full_auto", False otherwise.
             When enabled, agent.undo()/redo()/diff() become available.
-        sandbox: Optional SandboxConfig for execution isolation.
-            When level="full_auto" and sandbox is None, subprocess sandbox is auto-enabled.
+        sandbox: Optional SandboxConfig, carried for forward compatibility.
+            NOT IMPLEMENTED: nothing currently reads this field, and no sandbox
+            is auto-enabled at any autonomy level. To isolate execution, set it
+            on the agent itself -- Agent(sandbox=...) -- or run the whole
+            workflow remotely with AgentFlow(run_on="docker").
         snapshot_dir: Directory for snapshot storage. Defaults to ~/.praisonai/snapshots.
     
     Usage::
@@ -214,6 +217,13 @@ class AutonomyConfig:
                 f"Invalid autonomy level: {level!r}. "
                 f"Must be one of {sorted(VALID_AUTONOMY_LEVELS)}"
             )
+        verification_hooks = data.get("verification_hooks")
+        # Config parity: accept declarative `verification:` entries and map
+        # `command:`/`file:` shapes onto the two hook classes.
+        declarative = data.get("verification")
+        if declarative:
+            built = _build_verification_hooks(declarative)
+            verification_hooks = (verification_hooks or []) + built
         return cls(
             enabled=data.get("enabled", True),
             level=level,
@@ -224,7 +234,7 @@ class AutonomyConfig:
             observe=data.get("observe", False),
             completion_promise=data.get("completion_promise"),
             clear_context=data.get("clear_context", False),
-            verification_hooks=data.get("verification_hooks"),
+            verification_hooks=verification_hooks,
             track_changes=data.get("track_changes"),
             sandbox=data.get("sandbox"),
             snapshot_dir=data.get("snapshot_dir"),
@@ -234,6 +244,49 @@ class AutonomyConfig:
             budget_action=data.get("budget_action", "pause"),
         )
 
+
+def _build_verification_hooks(entries: List[Any]) -> List[Any]:
+    """Map declarative ``verification:`` config entries onto hook instances.
+
+    Each entry is a dict with either a ``command`` key (→
+    :class:`CommandVerificationHook`) or a ``file`` key (→
+    :class:`FileCheckHook`). Unknown shapes are skipped.
+
+    Example::
+
+        verification:
+          - {name: tests, command: "python -m pytest -q"}
+          - {name: changelog, file: CHANGELOG.md, non_empty: true}
+    """
+    from ..hooks.verification import CommandVerificationHook, FileCheckHook
+
+    hooks: List[Any] = []
+    for entry in entries or []:
+        if not isinstance(entry, dict):
+            continue
+        blocking = entry.get("blocking", True)
+        if "command" in entry:
+            hooks.append(CommandVerificationHook(
+                name=entry.get("name", "command"),
+                command=entry["command"],
+                cwd=entry.get("cwd"),
+                timeout=entry.get("timeout", 60.0),
+                blocking=blocking,
+            ))
+        elif "file" in entry:
+            hooks.append(FileCheckHook(
+                name=entry.get("name", "file"),
+                path=entry["file"],
+                exists=entry.get("exists", True),
+                non_empty=entry.get("non_empty", False),
+                contains=entry.get("contains"),
+                json_field=entry.get("json_field"),
+                equals=entry.get("equals"),
+                blocking=blocking,
+            ))
+    return hooks
+
+
 class AutonomySignal(str, Enum):
     """Signals detected from prompts for autonomy decisions.
     

src/praisonai-agents/praisonaiagents/agent/chat_mixin.py

@@ -2371,7 +2371,11 @@ Your Goal: {self.goal}"""
                     task_name=task_name,
                     task_description=task_description,
                     task_id=task_id,
-                    agent_name=getattr(self, 'name', 'assistant')
+                    agent_name=getattr(self, 'name', 'assistant'),
+                    # Durable re-injection target for deferred tool results:
+                    # background jobs completing after this call must land in the
+                    # persistent history a later turn replays, not the loop copy.
+                    deferred_history_sink=getattr(self, 'chat_history', None),
                 )
                 return final_response
             except ValueError as e:
@@ -2427,6 +2431,8 @@ Your Goal: {self.goal}"""
                     ]
                     or None
                 ),
+                # Durable re-injection target for deferred tool results (see above).
+                deferred_history_sink=getattr(self, 'chat_history', None),
             )
             return final_response
             
@@ -2895,6 +2901,14 @@ Your Goal: {self.goal}"""
                         'required' forces the LLM to call a tool before responding.
             ...other args...
         """
+        # Tools go to their sandbox before the first model call, once per
+        # agent. Placed on chat/achat/_start_stream rather than run/start
+        # because AgentTeam and AgentFlow call these three directly -- an
+        # agent used inside a team would otherwise run its tools locally.
+        if getattr(self, 'tools_run_on', None) is not None:
+            from .tools_placement import ensure_tools_placed
+            ensure_tools_placed(self)
+
         # Slash-command invocation: /skill-name [args] renders the skill
         # body before any backend/LLM call.
         prompt = self._resolve_skill_invocation(prompt)
@@ -3580,6 +3594,14 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
             attachments: Optional list of image/file paths that are ephemeral
                         (used for THIS turn only, NEVER stored in history).
         """
+        # Tools go to their sandbox before the first model call, once per
+        # agent. Placed on chat/achat/_start_stream rather than run/start
+        # because AgentTeam and AgentFlow call these three directly -- an
+        # agent used inside a team would otherwise run its tools locally.
+        if getattr(self, 'tools_run_on', None) is not None:
+            from .tools_placement import ensure_tools_placed
+            ensure_tools_placed(self)
+
         # Slash-command invocation: /skill-name [args] renders the skill body.
         prompt = self._resolve_skill_invocation(prompt)
 
@@ -4433,7 +4455,12 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                     # after-context aggregation) are fired once, guarded, inside
                     # execute_tool_async — no inline duplicate dispatch here.
                     # Pass the tools list to honor task-scoped tools
-                    result = await self.execute_tool_async(function_name, arguments, tools_override=tools)
+                    result = await self.execute_tool_async(
+                        function_name,
+                        arguments,
+                        tool_call_id=getattr(tool_call, "id", None),
+                        tools_override=tools,
+                    )
 
                     results.append(result)
                 except Exception as e:
@@ -4534,6 +4561,14 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
 
     def _start_stream(self, prompt: str, **kwargs) -> Generator[str, None, None]:
         """Own the durable context for the complete streaming generator."""
+        # Tools go to their sandbox before the first model call, once per
+        # agent. Placed on chat/achat/_start_stream rather than run/start
+        # because AgentTeam and AgentFlow call these three directly -- an
+        # agent used inside a team would otherwise run its tools locally.
+        if getattr(self, 'tools_run_on', None) is not None:
+            from .tools_placement import ensure_tools_placed
+            ensure_tools_placed(self)
+
         durable_context = None
         durable_token = None
         try:

src/praisonai-agents/praisonaiagents/agent/code_agent.py

@@ -59,7 +59,8 @@ class CodeAgent:
     def __init__(
         self,
         name: str = "CodeAgent",
-        llm: Optional[str] = None,
+        llm: Optional[str] = None,  # Deprecated alias for model=
+        model: Optional[str] = None,
         code: Optional[Union[bool, Dict, CodeConfig]] = None,
         instructions: Optional[str] = None,
         verbose: bool = True,
@@ -69,14 +70,16 @@ class CodeAgent:
         
         Args:
             name: Agent name
-            llm: LLM model (default: gpt-4o-mini)
+            llm: Deprecated alias for model=. Passing both raises TypeError.
+            model: Model to use (default: gpt-4o-mini)
             code: Code configuration (bool, dict, or CodeConfig)
             instructions: System instructions
             verbose: Enable verbose output
             **kwargs: Additional arguments
         """
+        from ..utils.model_alias import resolve_model_name
         self.name = name
-        self.llm = llm or "gpt-4o-mini"
+        self.llm = resolve_model_name(llm, model, "CodeAgent") or "gpt-4o-mini"
         self.instructions = instructions
         self.verbose = verbose
         

src/praisonai-agents/praisonaiagents/agent/durable.py

@@ -17,6 +17,7 @@ from ..runtime.journal import (
     KIND_MODEL_DECISION,
     KIND_TOOL_CALL,
     KIND_TOOL_RESULT,
+    KIND_VERIFICATION,
     RunJournal,
 )
 
@@ -277,6 +278,26 @@ class DurableRunContext:
         )
         self.finalize("failed")
 
+    def record_verification(self, payloads: List[Dict[str, Any]]) -> None:
+        """Append verification-gate results as durable ``verification`` events.
+
+        Each payload gets its own fresh ``seq`` so the events append cleanly
+        alongside the tool-call cursor without colliding on ``(run_id, seq,
+        kind)``. Best-effort: journalling never raises into the gate.
+        """
+        if self._closed:
+            return
+        with self._lock:
+            for payload in payloads:
+                seq = self._next_seq
+                self._next_seq += 1
+                try:
+                    self.journal.append(
+                        JournalEvent(self.run_id, seq, KIND_VERIFICATION, payload)
+                    )
+                except Exception:  # pragma: no cover - journalling is best-effort
+                    pass
+
     def wrap_sync(self, execute_tool_fn: Optional[Callable]) -> Optional[Callable]:
         if execute_tool_fn is None:
             return None

src/praisonai-agents/praisonaiagents/agent/embedding_agent.py

@@ -110,8 +110,8 @@ class EmbeddingAgent:
                 - EmbeddingConfig: Full configuration object
             verbose: Verbosity level for output
         """
-        if llm is None and model is not None:
-            llm = model
+        from ..utils.model_alias import resolve_model_name
+        llm = resolve_model_name(llm, model, "EmbeddingAgent")
         
         self.name = name or "EmbeddingAgent"
         self.instructions = instructions

src/praisonai-agents/praisonaiagents/agent/execution_location.py

@@ -0,0 +1,276 @@
+"""Describe, in plain English, where an agent's work actually happens.
+
+PraisonAI can run the thinking (LLM calls) and the tools in different places:
+this machine, a container, or a provider's cloud. Which one you get depends on
+``run_on=``, ``backend=`` and ``sandbox=``, and until now none of that was
+visible from the object -- ``repr(agent)`` was the bare default.
+
+This module is the single source of that answer, shared by ``Agent``,
+``AgentTeam`` and ``AgentFlow`` so all three describe themselves identically.
+
+Vocabulary note: the fields are ``thinks_on`` / ``tools_run_on``, never
+``loop``. "Loop" is already a public workflow primitive here
+(``praisonaiagents.workflows.loop()``), asyncio's event loop, and agent-loop
+jargon a beginner would not know -- three meanings for one word is exactly the
+problem this module exists to fix.
+"""
+
+from __future__ import annotations
+
+from typing import Any, Dict, Optional
+
+# Provider name -> how a human would say it out loud.
+_PLACE_WORDS = {
+    "local": "this machine",
+    "subprocess": "a separate process on this machine",
+    "sandlock": "a locked-down process on this machine",
+    "docker": "a Docker container",
+    "e2b": "an E2B cloud sandbox",
+    "modal": "a Modal cloud sandbox",
+    "daytona": "a Daytona cloud sandbox",
+    "flyio": "a Fly.io machine",
+    "tenki": "a Tenki cloud sandbox",
+    "novita": "a Novita cloud sandbox",
+    "ssh": "a remote machine over SSH",
+    "anthropic": "Anthropic's cloud",
+}
+
+HERE = "this machine"
+
+# Public sandbox aliases the SandboxManager collapses before it picks a backend
+# (see praisonaiagents/sandbox/manager.py). We must describe the backend that
+# actually runs, not the alias the user typed, or "native"/"local" would report
+# a place that never executes.
+_ALIASES = {"native": "sandlock", "local": "subprocess"}
+
+# ...but only for the SANDBOX registry. "local" resolves to two different
+# backends depending on which parameter carried it:
+#   run_in="local"        -> SandboxManager collapses it to subprocess:
+#                            scrubbed environment, blocked commands and paths
+#   tools_run_on="local"  -> the wrapper's LocalCompute: a plain shell in the
+#                            current directory, full environment inherited,
+#                            no security policy at all
+# Describing the second in the first's words overstates it, which is the exact
+# failure this module exists to prevent. One word, two backends -- so the
+# phrase has to depend on where the word came from.
+_COMPUTE_ONLY_WORDS = {"local": "a plain shell on this machine (no policy applied)"}
+
+# Tiers that separate a process but do NOT contain it. Verified: under the
+# subprocess backend, outbound network still works and ~/.ssh is readable even
+# though SecurityPolicy declares them blocked.
+# `native` is deliberately absent: it aliases sandlock, which applies Landlock,
+# seccomp, a scrubbed environment and deny-all networking. Listing it here made
+# tools_run_on="native" and tools_run_on="sandlock" -- the same backend --
+# disagree about whether they contain anything.
+_NOT_A_BOUNDARY = {"subprocess", "local"}
+
+
+def say_place(name: Any, *, via: str = "sandbox") -> str:
+    """Turn a place name into a phrase a non-developer can read.
+
+    ``via`` says which registry resolved the name -- ``"compute"`` for
+    ``tools_run_on=``, ``"sandbox"`` for ``run_in=``. It only matters for names
+    that mean different backends in each (see ``_COMPUTE_ONLY_WORDS``).
+    """
+    if name is None:
+        return HERE
+    if name is True:
+        return _PLACE_WORDS["subprocess"]
+    if not isinstance(name, str):
+        # A configured provider instance: use its own declared name if it has
+        # one. Normalise BEFORE the compute-only check, or a backend holding a
+        # LocalCompute object (provider_name="local") skips the branch and gets
+        # described in the subprocess backend's stronger words -- the exact
+        # overstatement this module exists to prevent.
+        name = getattr(name, "provider_name", None) or type(name).__name__
+    if via == "compute" and name.lower() in _COMPUTE_ONLY_WORDS:
+        return _COMPUTE_ONLY_WORDS[name.lower()]
+    key = str(name).lower()
+    key = _ALIASES.get(key, key)
+
+    known = _PLACE_WORDS.get(key)
+    if known:
+        return known
+
+    # A place contributed by another package can name itself; the bare name is
+    # the last resort rather than the only option, because the phrase table is
+    # a literal that cannot know about packages installed later.
+    try:
+        from ..managed._compute_bridge import contributed_display_name
+
+        declared = contributed_display_name(key)
+        if declared:
+            return declared
+    except Exception:
+        pass
+    return str(name)
+
+
+def _backend_places(backend: Any) -> Optional[Dict[str, str]]:
+    """Where does a managed ``backend=`` put the thinking and the tools?"""
+    if backend is None:
+        return None
+
+    # Ask the backend what it is, rather than guessing from its class name.
+    # Name-matching missed any backend that did not happen to contain "Hosted"
+    # or "Anthropic" -- a self-hosted docker backend reported that it was
+    # thinking on this machine while the whole loop ran in a container.
+    declared = getattr(backend, "provider_name", None)
+    if isinstance(declared, str):
+        try:
+            from .placement import managed_runtimes
+
+            if declared.lower() in managed_runtimes():
+                where = say_place(declared)
+                return {"thinks_on": where, "tools_run_on": where}
+        except Exception:
+            pass
+
+    cls = type(backend).__name__
+    # Hosted runtimes move the entire turn off this machine.
+    if "Hosted" in cls or "Anthropic" in cls:
+        provider = getattr(backend, "provider", None) or "anthropic"
+        where = say_place(provider)
+        return {"thinks_on": where, "tools_run_on": where}
+
+    # Local/sandboxed agents keep the loop here and may push tools out. A
+    # backend's compute object is a COMPUTE-registry provider, so it is read
+    # with the compute vocabulary: a LocalCompute here is the unrestricted
+    # shell, not the subprocess backend "local" collapses to under run_in=.
+    compute = getattr(backend, "_compute", None) or getattr(backend, "compute", None)
+    if compute is not None:
+        return {"thinks_on": HERE, "tools_run_on": say_place(compute, via="compute")}
+    return {"thinks_on": HERE, "tools_run_on": HERE}
+
+
+def describe(obj: Any, *, shared: bool = False) -> Dict[str, str]:
+    """Return ``{thinks_on, tools_run_on, code_runs_on?}`` for an object.
+
+    Never raises: a description of where things run must not be able to break
+    the thing it describes.
+    """
+    places: Dict[str, str] = {"thinks_on": HERE, "tools_run_on": HERE}
+    try:
+        from_backend = _backend_places(getattr(obj, "backend", None))
+        if from_backend:
+            places.update(from_backend)
+
+        where_tools = getattr(obj, "tools_run_on", None)
+        if where_tools is not None:
+            where = say_place(where_tools, via="compute")
+            places["tools_run_on"] = f"{where} (shared)" if shared else where
+
+        # run_on= puts the WHOLE agent on a managed runtime, so it moves the
+        # thinking as well. _backend_places() already reports that (run_on= is
+        # turned into a backend at construction); this only covers an object
+        # that carries the name without a built backend.
+        whole = getattr(obj, "run_on", None)
+        if whole is not None and not from_backend:
+            everywhere = say_place(whole)
+            places["thinks_on"] = everywhere
+            places["tools_run_on"] = everywhere
+
+        sandbox_config = getattr(obj, "sandbox_config", None)
+        if sandbox_config is not None:
+            places["code_runs_on"] = say_place(
+                getattr(sandbox_config, "sandbox_type", None)
+            )
+    except Exception:  # pragma: no cover - description must never break callers
+        pass
+    return places
+
+
+def _tools_staying_here(obj: Any) -> list:
+    """Names of the object's own tools, which do NOT move to the sandbox.
+
+    Only four tool names are bridged (``execute_command``, ``read_file``,
+    ``write_file``, ``list_files``); everything else the user passed keeps
+    running in this process, against this filesystem. The capability prompt
+    tells the model the sandbox is where things happen, so anything staying
+    behind has to be said out loud.
+
+    Reads the pre-attach tool list when a sandbox is already attached: by then
+    ``agent.tools`` has been rewritten and the originals live in the sandbox's
+    restore record.
+    """
+    try:
+        from ..managed.shared_compute import BRIDGED_TOOL_NAMES
+    except Exception:
+        BRIDGED_TOOL_NAMES = ("execute_command", "read_file", "write_file", "list_files")
+
+    tools = None
+    shared = getattr(obj, "_tools_sandbox", None)
+    if shared is not None:
+        for patched, original_tools, _ in getattr(shared, "_patched", []):
+            if patched is obj:
+                tools = original_tools
+                break
+    if tools is None:
+        tools = getattr(obj, "tools", None) or []
+
+    names = []
+    for tool in tools:
+        name = getattr(tool, "__name__", None) or getattr(tool, "name", None)
+        if isinstance(name, str) and name not in BRIDGED_TOOL_NAMES:
+            names.append(name)
+    return names
+
+
+def repr_fields(obj: Any, *, shared: bool = False) -> str:
+    """Render the places as ``key='value'`` pairs for a ``__repr__``."""
+    return ", ".join(f"{k}={v!r}" for k, v in describe(obj, shared=shared).items())
+
+
+def explain(obj: Any, *, shared: bool = False) -> str:
+    """Plain-sentence answer to "where does this actually run?".
+
+    Deliberately uses only words a non-engineer already knows -- no "loop",
+    "harness", "provision" or "runtime".
+    """
+    places = describe(obj, shared=shared)
+    lines = [
+        f"Thinking (the AI model calls) happens on {places['thinks_on']}.",
+        f"Tools run on {places['tools_run_on']}.",
+    ]
+
+    if shared and getattr(obj, "tools_run_on", None) is not None:
+        lines.append(
+            "Every step shares that same sandbox, so a file written by one step "
+            "is visible to the next."
+        )
+
+    staying = _tools_staying_here(obj)
+    if staying and places["tools_run_on"] != places["thinks_on"]:
+        listed = ", ".join(staying[:6]) + ("..." if len(staying) > 6 else "")
+        lines.append(
+            f"Your own tools ({listed}) still run on {HERE} -- only shell, file "
+            f"and code tools move. They read and write this machine's files."
+        )
+
+    if "code_runs_on" in places:
+        lines.append(
+            f"Code you run yourself with execute_code() runs on {places['code_runs_on']}."
+        )
+
+    # Say plainly when the chosen tier is not a security boundary.
+    raw = [
+        getattr(obj, "tools_run_on", None),
+        getattr(obj, "run_on", None),
+        getattr(getattr(obj, "sandbox_config", None), "sandbox_type", None),
+    ]
+    if any(isinstance(v, str) and v.lower() in _NOT_A_BOUNDARY for v in raw) or any(
+        v is True for v in raw
+    ):
+        lines.append(
+            "Note: a separate process is not a security boundary -- it can still "
+            "reach the network and read your files, and its blocked-command "
+            "list can be worked around by ordinary shell syntax. Use docker or a cloud "
+            "provider for untrusted code."
+        )
+
+    if places["thinks_on"] == HERE and places["tools_run_on"] == HERE:
+        lines.append(
+            "Nothing is isolated: tools you pass run in this program, with your "
+            "permissions."
+        )
+    return "\n".join(lines)

src/praisonai-agents/praisonaiagents/agent/execution_mixin.py

@@ -1456,6 +1456,18 @@ Write the complete compiled report:"""
                 after_results = await hook_runner.execute(
                     HookEvent.AFTER_TOOL, after_tool_input, target=function_name
                 )
+                # Honour a post-tool block so a redacted/denied result never
+                # reaches the model (mirrors the BEFORE_TOOL block path above).
+                if hook_runner.is_blocked(after_results):
+                    reason = hook_runner.get_blocking_reason(after_results)
+                    logging.warning(f"Tool {function_name} result blocked by AFTER_TOOL hook")
+                    return _record_async_loop_outcome(
+                        reason or f"Result of {function_name} was blocked by security policy."
+                    )
+                # Read back any rewritten/redacted tool_output so a plugin can
+                # scrub secrets before the result reaches the model.
+                if getattr(after_tool_input, "tool_output", result) is not result:
+                    result = after_tool_input.tool_output
                 extra_context = hook_runner.aggregate_context(after_results)
                 if extra_context:
                     if isinstance(result, str):
@@ -1507,6 +1519,7 @@ Write the complete compiled report:"""
                 run_id=getattr(self, '_current_run_id', 'unknown'),
                 session_id=getattr(self, '_session_id', None) or 'default',
                 tool_name=function_name,
+                metadata={"tool_call_id": tool_call_id},
             ),
         )
 
@@ -1528,6 +1541,7 @@ Write the complete compiled report:"""
 
     async def _execute_tool_async_with_retry(self, function_name: str, arguments: Dict[str, Any], tool_call_id: Optional[str] = None, tools_override: Optional[List] = None) -> Any:
         """Async tool execution with retry policy (middleware-agnostic core)."""
+        from ..errors import ToolExecutionError
 
         # Get retry policy (tool-level > agent-level > default)
         retry_policy = self._get_tool_retry_policy(function_name)
@@ -1544,22 +1558,39 @@ Write the complete compiled report:"""
                     # Skip retry for non-retryable errors (approval, permission, etc.).
                     # `retryable is False` covers async tool timeouts, whose executor
                     # work cannot be cancelled — retrying would duplicate side effects.
+                    # `loop_blocked` is a terminal loop-guard decision — retrying would
+                    # defeat the doom-loop protection this stop is meant to add.
                     if (result.get("approval_denied") or 
                         result.get("permission_denied") or 
                         result.get("approval_error") or
                         result.get("circuit_open") or
-                        result.get("retryable") is False):
+                        result.get("loop_blocked") or
+                        result.get("_praison_retryable") is False):
+                        result.pop("_praison_retryable", None)
                         return result
                     
                     # Determine error type for retry policy
                     error_type = self._classify_error_type(result, last_exception)
                     
-                    # Check if we should retry
-                    if not retry_policy.should_retry(error_type, attempt):
+                    # Check if we should retry. A tool that *raised* is flattened
+                    # into an error dict before it can be classified, so it always
+                    # arrives here as "unknown" — which no default retry_on
+                    # contains. The sync path escalates exactly those error types
+                    # to its own retry loop (tool_execution.py:853-866); honouring
+                    # the impl's explicit `retryable` flag is how the async path
+                    # reaches the same decision instead of never retrying at all.
+                    # `_praison_retryable` is a private control-plane key (see
+                    # _execute_tool_async_impl): a tool's own result may legitimately
+                    # contain "retryable" (common for HTTP-API wrappers) and must
+                    # never steer our retry loop.
+                    if (not retry_policy.should_retry(error_type, attempt)
+                            and result.get("_praison_retryable") is not True):
+                        result.pop("_praison_retryable", None)
                         return result
                     
                     # Don't retry on last attempt
                     if attempt == retry_policy.max_attempts - 1:
+                        result.pop("_praison_retryable", None)
                         return result
                     
                     # Emit retry hook event  
@@ -1573,6 +1604,11 @@ Write the complete compiled report:"""
                     # Success - return result
                     return result
                     
+            except ToolExecutionError:
+                # A loop-guard HALT (raised in _execute_tool_async_impl) is a
+                # terminal, non-retryable decision — let it propagate to stop the
+                # run instead of retrying or swallowing it into an error dict.
+                raise
             except Exception as e:
                 last_exception = e
                 # Determine if retryable
@@ -1604,6 +1640,7 @@ Write the complete compiled report:"""
 
     async def _execute_tool_async_impl(self, function_name: str, arguments: Dict[str, Any], tool_call_id: Optional[str] = None, tools_override: Optional[List] = None) -> Any:
         """Internal async tool execution implementation"""
+        from ..errors import ToolExecutionError
         try:
             logging.info(f"Executing async tool: {function_name} with arguments: {arguments}")
             
@@ -1651,6 +1688,29 @@ Write the complete compiled report:"""
                 logging.error(f"Function {function_name} not found in tools")
                 return {"error": f"Function {function_name} not found in tools"}
 
+            # Loop guard (pre-execution) — parity with the sync path in
+            # tool_execution.py. The async path previously had no loop-guard or
+            # doom-loop protection, so a repeatedly-failing tool could be hammered
+            # every turn via achat()/async workflows. check() is pure/sync, so it
+            # is safe to call here without awaiting.
+            loop_guard = None
+            if hasattr(self, '_ensure_loop_guard'):
+                loop_guard = self._ensure_loop_guard()
+                from ..escalation.loop_guard import GuardAction
+                decision = loop_guard.check(function_name, arguments, is_pre_execution=True)
+                if decision.action == GuardAction.BLOCK:
+                    logging.warning(f"Loop guard blocked {function_name}: {decision.message}")
+                    return {"error": f"[loop-guard] {decision.message}", "loop_blocked": True}
+                elif decision.action == GuardAction.HALT:
+                    raise ToolExecutionError(
+                        f"[loop-guard] {decision.message}",
+                        tool_name=function_name,
+                        agent_id=self.name,
+                        is_retryable=False,
+                    )
+                elif decision.action == GuardAction.WARN:
+                    logging.warning(f"Loop guard warning for {function_name}: {decision.message}")
+
             # Activate the tool-progress channel so tools running under the async
             # path can stream incremental output (emit_tool_progress) and the
             # built-in todo tool can publish live updates (emit_todo_update),
@@ -1723,14 +1783,33 @@ Write the complete compiled report:"""
                         # a duplicate invocation while the original keeps executing —
                         # duplicating DB writes / API calls / file mutations. Surface the
                         # timeout once instead of re-running an uncancellable side effect.
-                        return {
+                        # Fall through (not return) so the loop guard records this
+                        # timeout as a failure — repeated timeouts must accumulate
+                        # toward the BLOCK/HALT thresholds like any other failure.
+                        result = {
                             "error": f"Tool timed out after {tool_timeout}s",
                             "timeout": True,
-                            "retryable": False,
+                            "_praison_retryable": False,
                         }
                 else:
                     result = await _invoke()
-                
+
+                # Loop guard (post-execution) — record the outcome and surface a
+                # block/halt decision back to the model on this same turn, mirroring
+                # the sync path in tool_execution.py so repeated failures accumulate
+                # toward the BLOCK/HALT thresholds instead of retrying unbounded.
+                if loop_guard is not None:
+                    is_success = result is not None and not (isinstance(result, dict) and result.get('error'))
+                    loop_guard.record(function_name, arguments, is_success, result=result)
+                    decision = loop_guard.check(function_name, arguments, is_pre_execution=False)
+                    if decision.action.value in ("warn", "block", "halt"):
+                        if isinstance(result, str):
+                            result = f"{result}\n\n[loop-guard] {decision.message}"
+                        elif isinstance(result, dict):
+                            result["_loop_guard"] = {"message": decision.message, "action": decision.action.value}
+                        else:
+                            result = {"value": result, "_loop_guard": {"message": decision.message, "action": decision.action.value}}
+
                 # Ensure result is JSON serializable
                 logging.debug(f"Raw result from tool: {result}")
                 if result is None:
@@ -1744,24 +1823,61 @@ Write the complete compiled report:"""
 
             except Exception as e:
                 logging.error(f"Error executing {function_name}: {str(e)}", exc_info=True)
-                return {"error": f"Error executing {function_name}: {str(e)}"}
+                # Record the failed invocation so repeated identical failures
+                # accumulate toward the loop-guard BLOCK/HALT thresholds — a raised
+                # tool exception is a failure just like an error-dict result on the
+                # sync path. The next pre-execution check will then BLOCK/HALT.
+                if loop_guard is not None:
+                    loop_guard.record(function_name, arguments, False, result=None)
+                # Carry the retry verdict with the flattened exception. Without it
+                # the caller only sees an opaque error string, classifies it as
+                # "unknown" (never in RetryPolicy.retry_on) and gives up — while
+                # the sync path, which still has the exception object, retries.
+                # Mirror the sync path exactly: a ToolExecutionError already
+                # carries an explicit is_retryable verdict (e.g. a terminal
+                # non-retryable failure) and must be honoured verbatim; for any
+                # other exception apply the same terminal rule — programming
+                # errors are terminal, everything else may be retried.
+                if isinstance(e, ToolExecutionError):
+                    retryable = e.is_retryable
+                else:
+                    retryable = not isinstance(e, (ValueError, TypeError, AttributeError))
+                # Private control-plane key: a tool's own result may legitimately
+                # contain "retryable" (common for HTTP-API wrappers) and must not
+                # collide with the framework's verdict nor leak to the model.
+                return {
+                    "error": f"Error executing {function_name}: {str(e)}",
+                    "_praison_retryable": retryable,
+                }
 
+        except ToolExecutionError:
+            # A loop-guard HALT must propagate to stop the run, mirroring the sync
+            # path; do not swallow it into an error dict.
+            raise
         except Exception as e:
             logging.error(f"Error in execute_tool_async: {str(e)}", exc_info=True)
             return {"error": f"Error in execute_tool_async: {str(e)}"}
 
-    def launch(self, path: str = '/', port: int = 8000, host: str = '0.0.0.0', debug: bool = False, protocol: str = "http"):
+    def launch(self, path: str = '/', port: int = 8000, host: str = '127.0.0.1', debug: bool = False, protocol: str = "http"):
         """
         Launch the agent as an HTTP API endpoint or MCP server.
         
         This method now delegates to protocol-based launchers to follow
         the protocol-driven architecture. Heavy implementations (FastAPI/uvicorn)
         are only imported when needed through lazy loading.
-        
+
+        Request authorisation matches ``PraisonAIAgents.launch``: when
+        ``PRAISONAI_LAUNCH_AUTH_TOKEN`` is set, every HTTP route requires that
+        bearer token (401 otherwise). When it is unset and ``host`` is
+        non-loopback, a one-time token is generated and printed rather than
+        serving an unauthenticated endpoint on all interfaces.
+
         Args:
             path: API endpoint path (default: '/') for HTTP, or base path for MCP.
             port: Server port (default: 8000)
-            host: Server host (default: '0.0.0.0')
+            host: Server host (default: '127.0.0.1'; changed from '0.0.0.0' to
+                avoid binding all interfaces by default — pass '0.0.0.0'
+                explicitly to expose externally).
             debug: Enable debug mode (default: False)
             protocol: "http" to launch as FastAPI, "mcp" to launch as MCP server.
             
@@ -1785,6 +1901,8 @@ Write the complete compiled report:"""
         """
         global _server_started, _registered_agents, _shared_apps, _server_lock
 
+        from .launch_security import authorise_launch_request, resolve_launch_host
+
         # Try to import FastAPI dependencies - lazy loading
         try:
             import uvicorn
@@ -1809,7 +1927,15 @@ Write the complete compiled report:"""
             print("\nOr install all API dependencies with:")
             print("pip install 'praisonaiagents[api]'")
             return None
-                
+
+        # Fail-closed bind guard: refuse to serve keyless on a non-loopback host
+        # (auto-generates and prints a one-time token instead). Matches the
+        # serve/jobs precedent and keeps this path in lock-step with
+        # PraisonAIAgents.launch. Applied only after the dependency guard above
+        # so a launch that bails out (missing deps) cannot mutate the
+        # process-wide launch token and retroactively 401 a running endpoint.
+        host = resolve_launch_host(host)
+
         should_start = False
         with _server_lock:
             # Initialize port-specific collections if needed (once per port)
@@ -1862,6 +1988,8 @@ Write the complete compiled report:"""
             # Define the endpoint handler
             @_shared_apps[port].post(path)
             async def handle_agent_query(request: Request, query_data: Optional[AgentQuery] = None):
+                if not authorise_launch_request(request):
+                    raise HTTPException(status_code=401, detail="Unauthorized")
                 # Handle both direct JSON with query field and form data
                 if query_data is None:
                     try:

src/praisonai-agents/praisonaiagents/agent/launch_security.py

@@ -0,0 +1,88 @@
+"""Shared authorisation + bind guard for the ``launch()`` serving paths.
+
+Both ``Agent.launch()`` (``agent/execution_mixin.py``) and
+``PraisonAIAgents.launch()`` (``agents/agents.py``) expose a prompt-execution
+HTTP endpoint. They must enforce identical request authorisation and must not
+silently bind all interfaces without a token. Factoring the two helpers here
+keeps that policy in one place so the two launch paths cannot drift apart.
+
+Stdlib only — no heavy imports, no new public params.
+"""
+
+from __future__ import annotations
+
+import os
+import secrets as _secrets
+from typing import Optional
+
+__all__ = [
+    "launch_auth_token",
+    "authorise_launch_request",
+    "resolve_launch_host",
+]
+
+# Hosts that are safe to serve keyless on: loopback only.
+_LOOPBACK_HOSTS = frozenset({"127.0.0.1", "::1", "localhost"})
+
+# One auto-generated token per process, reused across launches so a later
+# non-loopback launch cannot mint a fresh token and lock out an endpoint that
+# is already serving with the previously generated one.
+_GENERATED_TOKEN: Optional[str] = None
+
+
+def launch_auth_token() -> Optional[str]:
+    """Return the configured launch bearer token, if any."""
+    return os.environ.get("PRAISONAI_LAUNCH_AUTH_TOKEN")
+
+
+def authorise_launch_request(request) -> bool:
+    """Return True if the request carries the configured bearer token.
+
+    When no token is configured this returns True (local-dev flow preserved).
+    Accepts either ``Authorization: Bearer <token>`` or ``X-Auth-Token``.
+    """
+    token = launch_auth_token()
+    if not token:
+        return True
+    auth = request.headers.get("Authorization", "")
+    if auth.startswith("Bearer ") and _secrets.compare_digest(auth[7:], token):
+        return True
+    supplied = request.headers.get("X-Auth-Token", "")
+    return bool(supplied) and _secrets.compare_digest(supplied, token)
+
+
+def _is_loopback(host: str) -> bool:
+    return host in _LOOPBACK_HOSTS
+
+
+def resolve_launch_host(host: str) -> str:
+    """Fail-closed bind guard, mirroring the ``serve``/jobs precedent.
+
+    * Token set → bind ``host`` unchanged (auth protects the endpoint).
+    * Token unset + loopback host → bind unchanged (local dev unchanged).
+    * Token unset + non-loopback host → auto-generate a token, export it via
+      ``PRAISONAI_LAUNCH_AUTH_TOKEN`` so every route enforces it, and print it
+      once. This closes the unauthenticated-exposure hole without refusing to
+      start.
+
+    Returns the host to bind (unchanged; the guard acts on the token, not the
+    host, so an explicit ``0.0.0.0`` is still honoured — just never keyless).
+
+    A generated token is reused for the lifetime of the process: repeated
+    non-loopback launches share the one token rather than each minting a fresh
+    one, so a later launch cannot invalidate an already-running endpoint's
+    clients.
+    """
+    if launch_auth_token() or _is_loopback(host):
+        return host
+
+    global _GENERATED_TOKEN
+    if _GENERATED_TOKEN is None:
+        _GENERATED_TOKEN = _secrets.token_urlsafe(32)
+        print(
+            f"🔐 launch() bound to non-loopback host {host!r} without "
+            f"PRAISONAI_LAUNCH_AUTH_TOKEN set. Generated a one-time bearer token "
+            f"(set PRAISONAI_LAUNCH_AUTH_TOKEN to override): {_GENERATED_TOKEN}"
+        )
+    os.environ["PRAISONAI_LAUNCH_AUTH_TOKEN"] = _GENERATED_TOKEN
+    return host

src/praisonai-agents/praisonaiagents/agent/ocr_agent.py

@@ -89,8 +89,8 @@ class OCRAgent:
             ocr: OCR configuration
             verbose: Verbosity level for output
         """
-        if llm is None and model is not None:
-            llm = model
+        from ..utils.model_alias import resolve_model_name
+        llm = resolve_model_name(llm, model, "OCRAgent")
         
         self.name = name or "OCRAgent"
         self.instructions = instructions

src/praisonai-agents/praisonaiagents/agent/placement.py

@@ -0,0 +1,211 @@
+"""One vocabulary for "where does this run?", shared by Agent, AgentFlow and AgentTeam.
+
+Two questions get asked about placement and they have different answers:
+
+``run_on=``
+    Where the **whole agent** runs -- the model calls, the loop, the tools, all
+    of it on someone else's machine. Only a *managed runtime* can answer this,
+    because running the loop remotely means something has to host the loop.
+
+``tools_run_on=``
+    Where **the tools** run. Thinking stays on this machine; only the shell,
+    the file writes and the code execution move. Any sandbox or compute place
+    can answer this, because all it has to do is run a command.
+
+Keeping the two words apart is the entire point. The old surface used one word
+(``run_on=``) on ``AgentFlow``/``AgentTeam`` for the tools-only meaning, another
+(``compute=``) on ``LocalAgent`` for exactly the same meaning, and a third
+(``backend=``) for the whole-agent meaning -- so the same concept had two names
+and one name had two concepts.
+
+Because the two questions have disjoint answer sets, a wrong pairing is a typo
+worth catching rather than a preference worth honouring. ``run_on="docker"``
+cannot mean anything: Docker does not host agent loops. It gets a ``TypeError``
+naming the parameter the user actually wanted, not a silent fallback.
+"""
+
+from __future__ import annotations
+
+from typing import Any, Literal, NamedTuple, Optional, Union
+
+# Runtimes that can host an entire agent loop. Resolved from the wrapper's
+# entry-point registry when it is installed so third-party backends work with
+# `run_on=` too; the literal is the floor, not the ceiling.
+_FALLBACK_MANAGED = ("anthropic",)
+
+# Static spellings so editors can autocomplete and type-checkers can reject a
+# typo at the call site rather than at the first model turn. The runtime
+# functions below stay the authority for validation -- these exist for tooling,
+# which cannot call a function to learn a set.
+# Keep in sync with _compute_bridge._PROVIDERS | _sandbox_adapter.SANDBOX_ONLY.
+ManagedRuntime = Literal["anthropic"]
+
+ToolPlace = Literal[
+    "local", "subprocess", "sandlock", "docker", "e2b",
+    "modal", "daytona", "flyio", "tenki", "novita", "ssh",
+]
+
+
+def managed_runtimes() -> tuple:
+    """Names that ``run_on=`` accepts -- places that host a whole agent."""
+    try:
+        from praisonai.integrations.backend_registry import get_backend_registry
+
+        names = tuple(get_backend_registry().list_all_names())
+        if names:
+            return tuple(sorted(set(names) | set(_FALLBACK_MANAGED)))
+    except Exception:
+        pass
+    return _FALLBACK_MANAGED
+
+
+def tool_places() -> list:
+    """Names that ``tools_run_on=`` accepts -- places that can run a command.
+
+    Includes the resolver's aliases. Validation runs before resolution, so
+    leaving them out rejected spellings the resolver handles happily:
+    ``tools_run_on="native"`` raised "not a known place" while
+    ``resolve_compute("native")`` returned a working sandlock adapter.
+    """
+    try:
+        from ..managed._compute_bridge import _ALIASES, available_providers
+
+        return sorted(set(available_providers()) | set(_ALIASES))
+    except Exception:
+        return ["local", "docker"]
+
+
+class Placement(NamedTuple):
+    """The resolved answer to "where does this run?"."""
+
+    whole: Optional[Any] = None   # managed runtime name/object, or None
+    tools: Optional[Any] = None   # tools place name/object, or None
+
+
+def _name_of(value: Any) -> Optional[str]:
+    """The comparable name of a placement value, or None if it is an object.
+
+    Objects (a built provider, a live ``SSHSandbox``) carry their own config and
+    are never second-guessed -- only bare strings get routed by name.
+    """
+    return value.lower() if isinstance(value, str) else None
+
+
+def _quote_list(names) -> str:
+    return ", ".join(repr(n) for n in names)
+
+
+def resolve_placement(
+    owner: str,
+    *,
+    run_on: Any = None,
+    tools_run_on: Any = None,
+    backend: Any = None,
+    supports_run_on: bool = True,
+) -> Placement:
+    """Validate placement arguments and return where things should run.
+
+    ``owner`` is the class name, used only to make errors read naturally.
+    ``supports_run_on`` is False for AgentFlow/AgentTeam, which can place tools
+    but cannot (yet) hand a whole multi-agent run to a managed runtime.
+
+    Raises:
+        TypeError: on any combination that names two different places for the
+            same thing, or a place that cannot do the job asked of it.
+    """
+    managed = managed_runtimes()
+    places = tool_places()
+
+    # ── a place that cannot do the job asked of it ───────────────────────────
+    run_on_name = _name_of(run_on)
+    if run_on_name is not None and run_on_name not in managed:
+        if run_on_name in places:
+            raise TypeError(
+                f"{owner}(run_on={run_on!r}) is not valid: run_on= places the "
+                f"whole agent -- model calls, loop and tools -- on a managed "
+                f"runtime, and {run_on_name!r} runs commands but cannot host an "
+                f"agent loop.\n"
+                f"  To run only the tools there:  {owner}(tools_run_on={run_on!r})\n"
+                f"  Runtimes that host a whole agent: {_quote_list(managed)}"
+            )
+        raise TypeError(
+            f"{owner}(run_on={run_on!r}) is not a known managed runtime.\n"
+            f"  Runtimes that host a whole agent: {_quote_list(managed)}\n"
+            f"  Places that can run tools:        {_quote_list(places)} "
+            f"(pass those as tools_run_on=)"
+        )
+
+    tools_name = _name_of(tools_run_on)
+    if tools_name is not None and tools_name in managed and tools_name not in places:
+        raise TypeError(
+            f"{owner}(tools_run_on={tools_run_on!r}) is not valid: {tools_name!r} "
+            f"hosts an entire agent, not individual tools.\n"
+            f"  Did you mean:  {owner}(run_on={tools_run_on!r})"
+        )
+
+    # A typo here used to survive construction and only surface on the first
+    # model turn -- after the prompt was built and, for a real run, after the
+    # API spend. run_on= validates at the call site and YAML validates at load
+    # time; this is the same standard for the third spelling.
+    if tools_name is not None and tools_name not in places:
+        raise TypeError(
+            f"{owner}(tools_run_on={tools_run_on!r}) is not a known place.\n"
+            f"  Places that can run tools: {_quote_list(places)}\n"
+            f"  Runtimes that host a whole agent: {_quote_list(managed)} "
+            f"(pass those as run_on=)"
+        )
+
+    # Names that cannot work as a bare string, because the connection details
+    # live in the object. Catch it now rather than at first use.
+    if tools_name is not None:
+        try:
+            from ..managed._sandbox_adapter import NEEDS_INSTANCE
+        except Exception:
+            NEEDS_INSTANCE = {}
+        if tools_name in NEEDS_INSTANCE:
+            raise TypeError(
+                f"{owner}(tools_run_on={tools_run_on!r}) needs more than a name: "
+                f"{NEEDS_INSTANCE[tools_name]}"
+            )
+
+    # ── two names for the same thing ─────────────────────────────────────────
+    if run_on is not None and backend is not None:
+        raise TypeError(
+            f"{owner}(run_on=..., backend=...) sets the agent's runtime twice. "
+            f"run_on={run_on!r} is the short form of backend=HostedAgent("
+            f"provider={run_on!r}); pass one or the other.\n"
+            f"  Use backend= only when you need to configure the runtime "
+            f"(model, system prompt, tools)."
+        )
+
+    if run_on is not None and tools_run_on is not None:
+        raise TypeError(
+            f"{owner}(run_on={run_on!r}, tools_run_on={tools_run_on!r}) points "
+            f"the tools at two machines. run_on= already runs everything -- "
+            f"including tools -- on {run_on!r}.\n"
+            f"  Whole agent remote:  {owner}(run_on={run_on!r})\n"
+            f"  Only tools remote:   {owner}(tools_run_on={tools_run_on!r})"
+        )
+
+    if backend is not None and tools_run_on is not None:
+        raise TypeError(
+            f"{owner}(backend=..., tools_run_on={tools_run_on!r}) points the "
+            f"tools at two machines: backend= runs the whole agent (tools "
+            f"included) on its managed runtime, so the tools cannot also run on "
+            f"{tools_run_on!r}.\n"
+            f"  Drop tools_run_on= to keep the hosted runtime, or drop backend= "
+            f"to keep local thinking with remote tools."
+        )
+
+    # ── a class that cannot answer this question ─────────────────────────────
+    if run_on is not None and not supports_run_on:
+        raise TypeError(
+            f"{owner}(run_on={run_on!r}) is not supported: run_on= hands one "
+            f"agent's whole loop to a managed runtime, and a {owner} "
+            f"orchestrates several agents locally.\n"
+            f"  To run every step's tools in one shared sandbox:\n"
+            f"      {owner}(tools_run_on={run_on!r})\n"
+            f"  To host a single agent's loop, put run_on= on that Agent."
+        )
+
+    return Placement(whole=run_on, tools=tools_run_on)

src/praisonai-agents/praisonaiagents/agent/realtime_agent.py

@@ -57,7 +57,8 @@ class RealtimeAgent:
     def __init__(
         self,
         name: str = "RealtimeAgent",
-        llm: Optional[str] = None,
+        llm: Optional[str] = None,  # Deprecated alias for model=
+        model: Optional[str] = None,
         realtime: Optional[Union[bool, Dict, RealtimeConfig]] = None,
         instructions: Optional[str] = None,
         verbose: bool = True,
@@ -67,14 +68,16 @@ class RealtimeAgent:
         
         Args:
             name: Agent name
-            llm: LLM model (default: gpt-4o-realtime-preview)
+            llm: Deprecated alias for model=. Passing both raises TypeError.
+            model: Model to use (default: gpt-4o-realtime-preview)
             realtime: Realtime configuration (bool, dict, or RealtimeConfig)
             instructions: System instructions
             verbose: Enable verbose output
             **kwargs: Additional arguments
         """
+        from ..utils.model_alias import resolve_model_name
         self.name = name
-        self.llm = llm or "gpt-4o-realtime-preview"
+        self.llm = resolve_model_name(llm, model, "RealtimeAgent") or "gpt-4o-realtime-preview"
         self.instructions = instructions
         self.verbose = verbose
         

src/praisonai-agents/praisonaiagents/agent/sandbox_mixin.py

@@ -52,19 +52,61 @@ class SandboxMixin:
         
         return self._sandbox_manager
     
+
+    def _manager_for(self, run_in: Optional[Union[bool, str, 'SandboxConfig']]):
+        """Return the sandbox manager for a ``run_in=`` argument.
+
+        ``run_in=`` names where a single ``execute_code()`` call runs, so the
+        place lives on the call that uses it rather than on the constructor.
+        ``Agent(sandbox=...)`` still sets the default for every call.
+
+        Managers are cached per place: a fresh ``SandboxManager`` per call would
+        start and tear down a container each time, which for Docker is seconds
+        of latency and for any backend means the previous call's files are gone.
+        """
+        if run_in is None:
+            if not self.has_sandbox:
+                raise RuntimeError(
+                    "No sandbox configured. Either name a place on the call --\n"
+                    "    agent.execute_code_sync(code, run_in='docker')\n"
+                    "or set a default for every call with Agent(sandbox='docker')."
+                )
+            return self.get_sandbox_manager()
+
+        from ..sandbox import SandboxConfig, SandboxManager
+
+        if run_in is True:
+            config = SandboxConfig.subprocess()
+        elif isinstance(run_in, str):
+            config = SandboxConfig(sandbox_type=run_in)
+        else:
+            config = run_in
+
+        key = getattr(config, "sandbox_type", None) or repr(config)
+        cache = getattr(self, "_sandbox_managers", None)
+        if cache is None:
+            cache = self._sandbox_managers = {}
+        if key not in cache:
+            cache[key] = SandboxManager(config)
+        return cache[key]
+
     async def execute_code(
         self,
         code: str,
         language: str = "python",
         check_security: bool = True,
+        run_in: Optional[Union[bool, str, 'SandboxConfig']] = None,
         **kwargs
     ) -> 'SandboxResult':
-        """Execute code safely in configured sandbox.
+        """Execute code safely in a sandbox.
         
         Args:
             code: Code to execute
             language: Programming language (python, bash, etc.)
             check_security: Whether to run security pre-checks
+            run_in: Where to run this call -- "docker", "sandlock",
+                "subprocess", a SandboxConfig, or True for the default local
+                sandbox. Overrides Agent(sandbox=...) for this call only.
             **kwargs: Additional arguments passed to sandbox.execute()
             
         Returns:
@@ -73,31 +115,39 @@ class SandboxMixin:
         Raises:
             RuntimeError: If no sandbox is configured
         """
-        if not self.has_sandbox:
-            raise RuntimeError(
-                "No sandbox configured. Set sandbox=True or provide SandboxConfig to Agent()"
-            )
+        manager = self._manager_for(run_in)
         
         # Security pre-check if enabled
+        security_warnings = None
         if check_security:
             from ..sandbox import check_code_safety, format_warnings
-            warnings = check_code_safety(code, language)
-            if warnings:
-                warning_text = format_warnings(warnings)
+            security_warnings = check_code_safety(code, language)
+            if security_warnings:
+                warning_text = format_warnings(security_warnings)
                 if self.verbose:
                     logger.warning(f"Security warnings for code execution:\n{warning_text}")
-                
-                # Store warnings in result metadata
-                kwargs.setdefault('metadata', {})['security_warnings'] = warnings
-        
-        manager = self.get_sandbox_manager()
-        return await manager.run_code(code, language=language, **kwargs)
+
+        result = await manager.run_code(code, language=language, **kwargs)
+
+        # Attach warnings to the result rather than forwarding a `metadata`
+        # kwarg: SandboxProtocol.execute() has no such parameter, so passing it
+        # through raised TypeError for any code that tripped a warning -- i.e.
+        # sandboxed execution failed exactly when the check mattered most.
+        if security_warnings:
+            try:
+                existing = getattr(result, "metadata", None) or {}
+                existing["security_warnings"] = security_warnings
+                result.metadata = existing
+            except Exception:  # pragma: no cover - result may be immutable
+                logger.debug("Could not attach security warnings to sandbox result")
+        return result
     
     def execute_code_sync(
         self,
         code: str,
         language: str = "python", 
         check_security: bool = True,
+        run_in: Optional[Union[bool, str, 'SandboxConfig']] = None,
         **kwargs
     ) -> 'SandboxResult':
         """Synchronous wrapper for execute_code.
@@ -112,12 +162,15 @@ class SandboxMixin:
             SandboxResult with execution details
         """
         from ..approval.utils import run_coroutine_safely
-        return run_coroutine_safely(self.execute_code(code, language, check_security, **kwargs))
+        return run_coroutine_safely(
+            self.execute_code(code, language, check_security, run_in=run_in, **kwargs)
+        )
     
     async def run_shell_command(
         self,
         command: Union[str, list],
         check_security: bool = True,
+        run_in: Optional[Union[bool, str, 'SandboxConfig']] = None,
         **kwargs
     ) -> 'SandboxResult':
         """Run a shell command in the sandbox.
@@ -130,8 +183,7 @@ class SandboxMixin:
         Returns:
             SandboxResult with execution details
         """
-        if not self.has_sandbox:
-            raise RuntimeError("No sandbox configured")
+        manager = self._manager_for(run_in)
         
         # Convert command to string for security checking
         if isinstance(command, list):
@@ -148,7 +200,7 @@ class SandboxMixin:
                 if self.verbose:
                     logger.warning(f"Security warnings for command:\n{warning_text}")
         
-        async with self.get_sandbox_manager() as sandbox:
+        async with manager as sandbox:
             return await sandbox.run_command(command, **kwargs)
     
     def get_sandbox_status(self) -> Dict[str, Any]:
@@ -230,5 +282,15 @@ class SandboxMixin:
                 return result.stdout or result.output
             else:
                 return f"Error: {result.error or result.stderr}"
-        
-        return [execute_python_code, execute_shell_command]
+
+        # Tag as framework-generated so clone_for_channel() can drop these
+        # source-bound wrappers and regenerate clone-owned ones. Without the
+        # tag, a clone would inherit tools closing over the *source* agent and
+        # silently share its SandboxManager.
+        generated = [execute_python_code, execute_shell_command]
+        for _t in generated:
+            try:
+                _t._praison_sandbox_tool = True
+            except Exception:  # pragma: no cover - tool object may be immutable
+                pass
+        return generated

src/praisonai-agents/praisonaiagents/agent/skill_review.py

@@ -52,14 +52,29 @@ class SkillReviewMixin:
         return policy
 
     def _build_skill_review_tool(self):
-        """Return the restricted toolset for the review turn (skill_manage only)."""
+        """Return the restricted toolset for the review turn.
+
+        Always exposes the single skill-mutation tool (``skill_manage``). When
+        the agent has memory enabled, ``store_memory`` is added so the same
+        guarded pass can also persist a durable fact / user-model detail
+        (issue #4030) — closing the learning loop for memory, not just skills,
+        without a second review turn or a parallel knob. The set stays
+        restricted: never the agent's full toolset.
+        """
+        tools = []
         try:
             from ..tools.skill_tools import create_skill_tools
             skill_tools = create_skill_tools()
-            return [skill_tools.skill_manage]
+            tools.append(skill_tools.skill_manage)
         except Exception as e:  # pragma: no cover - defensive
             logger.debug("Could not build skill review tool: %s", e, exc_info=True)
-            return []
+        if getattr(self, "memory", None):
+            try:
+                from ..tools.memory import store_memory
+                tools.append(store_memory)
+            except Exception as e:  # pragma: no cover - defensive
+                logger.debug("Could not build memory review tool: %s", e, exc_info=True)
+        return tools
 
     def _skill_review_log_context(self):
         """Return (agent_name, session_id) for swallowed-failure warnings."""

src/praisonai-agents/praisonaiagents/agent/tool_execution.py

@@ -22,6 +22,13 @@ from ..config.feature_configs import DEFAULT_TOOL_OUTPUT_LIMIT
 
 logger = logging.getLogger(__name__)
 
+# Upper bound on how many timed-out (still-blocked) tool executors a single
+# agent may retire before it stops recycling and falls back to reusing the
+# existing bounded pool. This keeps the "fresh worker after a timeout"
+# throughput win while capping leaked threads from pathological repeated
+# timeouts so they cannot exhaust process resources.
+_MAX_ORPHANED_TOOL_EXECUTORS = 4
+
 if TYPE_CHECKING:
     pass
 
@@ -367,6 +374,12 @@ class ToolExecutionMixin:
 
         return resolve_tool_names(tool_names)
 
+    def _resolve_tools_list(self, tools):
+        """Resolve any tool-name strings inside a mixed ``tools=`` list."""
+        from ..tools.resolver import resolve_tools_list
+
+        return resolve_tools_list(tools)
+
     def _cast_arguments(self, func, arguments):
         """Cast arguments to their expected types based on function signature."""
         if not callable(func) or not arguments:
@@ -482,6 +495,7 @@ class ToolExecutionMixin:
                 run_id=getattr(self, '_current_run_id', 'unknown'),
                 session_id=getattr(self, '_session_id', None) or 'default',
                 tool_name=function_name,
+                metadata={"tool_call_id": tool_call_id},
             ),
         )
 
@@ -739,26 +753,21 @@ class ToolExecutionMixin:
             if blocked_result is not None:
                 result = blocked_result
             else:
-                # Apply tool retry logic with exponential backoff
-                execution_config = getattr(self, '_execution_config', None)
-                if execution_config is None:
-                    # Fall back to reading individual config attributes for backward compatibility
-                    max_retry_limit = getattr(self, 'max_retry_limit', 2)
-                    retry_initial_delay = 1.0
-                    retry_backoff_factor = 2.0
-                    retry_jitter = 0.1
-                else:
-                    max_retry_limit = execution_config.max_retry_limit
-                    retry_initial_delay = execution_config.retry_initial_delay
-                    retry_backoff_factor = execution_config.retry_backoff_factor
-                    retry_jitter = execution_config.retry_jitter
-                
+                # Apply tool retry logic with exponential backoff.
+                # ONE vocabulary: the effective RetryPolicy caps how many times
+                # the tool body may run, whichever of the two loops does the
+                # retrying, and owns the backoff. ExecutionConfig.max_retry_limit
+                # / retry_* are translated into it (see
+                # _effective_tool_retry_policy) rather than run as a second,
+                # independent budget in different units.
+                retry_policy = self._effective_tool_retry_policy(function_name)
+                max_attempts = retry_policy.max_attempts
+
                 result = None
                 last_exception = None
-                
-                # max_retry_limit is the number of retries (not total attempts)
-                # So total attempts = 1 (initial) + max_retry_limit (retries)
-                for attempt in range(1, max_retry_limit + 2):
+
+                # max_attempts is the total number of tool-body runs.
+                for attempt in range(1, max_attempts + 1):
                     try:
                         # P8/G11: Apply tool timeout if configured
                         tool_timeout = getattr(self, '_tool_timeout', None)
@@ -787,7 +796,7 @@ class ToolExecutionMixin:
                                         return self._execute_tool_with_circuit_breaker(function_name, arguments)
 
                                 # Use reusable executor to prevent resource leaks
-                                if not hasattr(self, '_tool_executor'):
+                                if not hasattr(self, '_tool_executor') or self._tool_executor is None:
                                     self._tool_executor = concurrent.futures.ThreadPoolExecutor(
                                         max_workers=2, thread_name_prefix=f"tool-{self.name}"
                                     )
@@ -800,66 +809,94 @@ class ToolExecutionMixin:
                                         _progress_active.clear()
                                     future.cancel()
                                     logging.warning(f"Tool {function_name} timed out after {tool_timeout}s")
-                                    result = {"error": f"Tool timed out after {tool_timeout}s", "timeout": True}
+                                    # Tag as an OUTER wall-clock timeout: unlike an
+                                    # inner-loop timeout (already retried to
+                                    # exhaustion), the body ran once here and the
+                                    # inner budget never saw it, so this one still
+                                    # earns an outer retry pass under the single budget.
+                                    result = {"error": f"Tool timed out after {tool_timeout}s", "timeout": True, "_outer_timeout": True}
+                                    # The timed-out thread cannot be reclaimed. Retire this
+                                    # executor so the next call gets a fresh worker instead of
+                                    # queuing behind a permanently stuck one — but bound how many
+                                    # orphaned (still-blocked) threads we allow to accumulate:
+                                    # once the cap is reached we stop recycling and keep reusing
+                                    # the existing pool so repeated timeouts can't exhaust process
+                                    # resources with an unbounded number of leaked threads.
+                                    orphaned = getattr(self, '_tool_executor_orphaned', 0)
+                                    if orphaned < _MAX_ORPHANED_TOOL_EXECUTORS:
+                                        self._tool_executor.shutdown(wait=False)
+                                        self._tool_executor = None
+                                        self._tool_executor_orphaned = orphaned + 1
                         else:
                             with tool_progress_channel(_progress_sink), with_injection_context(state):
                                 result = self._execute_tool_with_circuit_breaker(function_name, arguments)
                         
                         # Check if the result indicates a retryable error
                         if isinstance(result, dict) and result.get("error"):
-                            # Check if this is a circuit breaker error (always retryable)
-                            if result.get("circuit_open"):
-                                raise ToolExecutionError(
-                                    result["error"],
-                                    tool_name=function_name,
-                                    agent_id=self.name,
-                                    is_retryable=True,
-                                )
-                            # Check if this is a timeout error (retryable)
-                            elif result.get("timeout"):
-                                raise ToolExecutionError(
-                                    result["error"],
-                                    tool_name=function_name,
-                                    agent_id=self.name,
-                                    is_retryable=True,
-                                )
-                            # For other error dicts: approval/permission denials are legitimate
-                            # non-retryable outcomes; everything else represents a tool failure
-                            # that should engage the outer retry/backoff loop.
-                            elif result.get("approval_denied") or result.get("permission_denied") or result.get("approval_error") or result.get("policy_denied") or result.get("guardrail_denied"):
+                            # Approval/permission/policy/guardrail denials are
+                            # legitimate non-retryable outcomes: stop immediately.
+                            if result.get("approval_denied") or result.get("permission_denied") or result.get("approval_error") or result.get("policy_denied") or result.get("guardrail_denied"):
                                 break
-                            else:
-                                # Avoid compounding with the inner retry loop in
-                                # _execute_tool_with_circuit_breaker: error types it already
-                                # retries (e.g. timeout/rate_limit/connection_error) are
-                                # exhausted by the time they reach here, so escalating them as
-                                # retryable would re-run the entire inner loop from scratch on
-                                # every outer attempt. Only escalate error types the inner loop
-                                # does NOT retry (e.g. "unknown") so they get one outer pass.
-                                error_type = self._classify_error_type(result, None)
-                                inner_policy = self._get_tool_retry_policy(function_name)
-                                is_retryable = error_type not in inner_policy.retry_on
-                                raise ToolExecutionError(
-                                    result.get("error", f"Tool '{function_name}' failed"),
-                                    tool_name=function_name,
-                                    agent_id=self.name,
-                                    is_retryable=is_retryable,
-                                )
+                            # ONE budget: the inner loop
+                            # (_execute_tool_with_circuit_breaker) already retries
+                            # its retry_on error types (timeout/rate_limit/
+                            # connection_error) to exhaustion before returning here,
+                            # so re-escalating them as retryable would restart the
+                            # whole inner budget on every outer attempt and run the
+                            # tool up to max_attempts² times.
+                            error_type = self._classify_error_type(result, None)
+                            inner_policy = self._get_tool_retry_policy(function_name)
+                            # Only escalate to an outer retry pass when the failure is
+                            # a transient the inner loop did NOT already re-drive:
+                            #   • _outer_timeout — this loop's own wall-clock guard
+                            #     tripped; the body ran once and the inner budget
+                            #     never saw it, so it earns one outer pass.
+                            #   • circuit_open — returned by the inner loop WITHOUT a
+                            #     body run, so retrying costs no duplicate side effect.
+                            #   • _praison_retryable — a raised exception was flattened
+                            #     into this error dict (see _execute_tool_impl); honour
+                            #     that verdict, but ONLY for error types the inner loop
+                            #     does not itself retry, or inner×outer would run the
+                            #     body max_attempts² times.
+                            # A plain error dict the tool body *returned* carries no tag
+                            # — it is the tool's own reported outcome, not a transient
+                            # the framework must re-drive. The body already ran to
+                            # completion, so re-running it would duplicate its side
+                            # effects (charge money, send a message, write a row) for a
+                            # decision the tool already made. If a user genuinely wants
+                            # such an outcome retried, its error type belongs in the
+                            # RetryPolicy.retry_on the inner loop honours — not a word
+                            # in the tool's own payload. Surface it once.
+                            is_retryable = bool(
+                                result.get("_outer_timeout")
+                                or result.get("circuit_open")
+                                or (result.get("_praison_retryable") is True
+                                    and error_type not in inner_policy.retry_on)
+                            )
+                            # Strip the private control-plane tag before it can reach
+                            # the model or be re-surfaced as the tool's payload.
+                            result.pop("_praison_retryable", None)
+                            raise ToolExecutionError(
+                                result.get("error", f"Tool '{function_name}' failed"),
+                                tool_name=function_name,
+                                agent_id=self.name,
+                                is_retryable=is_retryable,
+                            )
                         else:
                             # Success path - return the result
                             break
                             
                     except ToolExecutionError as e:
                         last_exception = e
-                        # Only retry if the error is marked as retryable and we have retries left
-                        # attempt starts at 1, so (attempt - 1) gives us the retry count
-                        if not e.is_retryable or (attempt - 1) >= max_retry_limit:
+                        # Only retry if the error is marked as retryable and the
+                        # policy still has attempts left.
+                        if not e.is_retryable or attempt >= max_attempts:
                             raise e
-                        
-                        # Calculate delay for exponential backoff
-                        delay = BackoffPolicy.delay(attempt, retry_initial_delay, retry_backoff_factor, retry_jitter)
+
+                        # Calculate delay for exponential backoff (policy owns it)
+                        delay = retry_policy.get_delay_ms(attempt - 1) / 1000.0
                         logging.warning(
-                            f"Tool {function_name} failed (attempt {attempt}/{max_retry_limit + 1}): {e}. "
+                            f"Tool {function_name} failed (attempt {attempt}/{max_attempts}): {e}. "
                             f"Retrying in {delay:.2f}s..."
                         )
                         time.sleep(delay)
@@ -875,15 +912,14 @@ class ToolExecutionMixin:
                             is_retryable=is_retryable,
                         )
                         last_exception = tool_error
-                        
-                        # attempt starts at 1, so (attempt - 1) gives us the retry count
-                        if not is_retryable or (attempt - 1) >= max_retry_limit:
+
+                        if not is_retryable or attempt >= max_attempts:
                             raise tool_error from e
-                        
-                        # Calculate delay for exponential backoff
-                        delay = BackoffPolicy.delay(attempt, retry_initial_delay, retry_backoff_factor, retry_jitter)
+
+                        # Calculate delay for exponential backoff (policy owns it)
+                        delay = retry_policy.get_delay_ms(attempt - 1) / 1000.0
                         logging.warning(
-                            f"Tool {function_name} failed (attempt {attempt}/{max_retry_limit + 1}): {e}. "
+                            f"Tool {function_name} failed (attempt {attempt}/{max_attempts}): {e}. "
                             f"Retrying in {delay:.2f}s..."
                         )
                         time.sleep(delay)
@@ -1130,6 +1166,20 @@ class ToolExecutionMixin:
                 )
                 after_tool_results = self._hook_runner.execute_sync(HookEvent.AFTER_TOOL, after_tool_input, target=function_name)
 
+                # Honour a post-tool block: a plugin/hook can deny the result
+                # (e.g. secret/PII detected) so the original output never
+                # reaches the model. Mirrors the BEFORE_TOOL block path above.
+                if self._hook_runner.is_blocked(after_tool_results):
+                    reason = self._hook_runner.get_blocking_reason(after_tool_results)
+                    logging.warning(f"Tool {function_name} result blocked by AFTER_TOOL hook")
+                    return reason or f"Result of {function_name} was blocked by security policy."
+
+                # Read back any rewritten/redacted tool_output so a plugin can
+                # scrub secrets before the result reaches the model. The bridge
+                # mutates ``after_tool_input.tool_output`` in place.
+                if getattr(after_tool_input, "tool_output", result) is not result:
+                    result = after_tool_input.tool_output
+
                 # Surface any additional_context returned by AFTER_TOOL hooks
                 # back to the model by appending it to the tool result (mirrors
                 # the loop-guard injection pattern below). Without this the
@@ -1961,6 +2011,12 @@ class ToolExecutionMixin:
         if not decision.approved:
             error_msg = f"Tool execution denied: {decision.reason}"
             logging.warning(error_msg)
+            feedback = getattr(decision, "feedback", None)
+            if feedback:
+                error_msg = (
+                    f"Tool '{function_name}' was denied by the user. "
+                    f"Do not retry it as-is. User guidance: {feedback}"
+                )
             return {"error": error_msg, "approval_denied": True}
         
         if decision.modified_args:
@@ -1994,6 +2050,12 @@ class ToolExecutionMixin:
         if not decision.approved:
             error_msg = f"Tool execution denied: {decision.reason}"
             logging.warning(error_msg)
+            feedback = getattr(decision, "feedback", None)
+            if feedback:
+                error_msg = (
+                    f"Tool '{function_name}' was denied by the user. "
+                    f"Do not retry it as-is. User guidance: {feedback}"
+                )
             return {"error": error_msg, "approval_denied": True}
         
         if decision.modified_args:
@@ -2518,8 +2580,20 @@ class ToolExecutionMixin:
                             f"{error_msg} Expected parameters for '{function_name}' — "
                             f"required: {schema['required']}, optional: {schema['optional']}."
                         )
-                        return {"error": error_msg, "expected_parameters": schema}
-                return {"error": error_msg}
+                        # An argument-binding error is a programming error — the same
+                        # call will always fail — so mark it terminal and never retry.
+                        return {"error": error_msg, "expected_parameters": schema, "_praison_retryable": False}
+                # A tool that *raised* is flattened into an error dict here, which
+                # erases the exception type the retry policy needs. Carry the same
+                # verdict the exception paths use so the outer loop can retry a
+                # transient failure without re-running a tool that merely *returned*
+                # an error dict of its own. Programming errors are terminal;
+                # everything else may be retried. Private key: never leaks to the
+                # model and cannot collide with a tool's own "retryable" field.
+                return {
+                    "error": error_msg,
+                    "_praison_retryable": not isinstance(e, (ValueError, TypeError, AttributeError)),
+                }
 
         # Unresolved: return a corrective, model-readable message so the model can
         # retry with a valid name instead of repeating the same mistake.
@@ -2908,15 +2982,33 @@ class ToolExecutionMixin:
 
     def _get_tool_retry_policy(self, tool_name):
         """Get retry policy for a tool (tool-level > agent-level > default).
-        
+
+        Kept as the historical name; both retry loops now go through
+        ``_effective_tool_retry_policy`` so they share one policy object.
+
         Args:
             tool_name: Name of the tool to get retry policy for
             
         Returns:
             RetryPolicy instance
         """
+        return self._effective_tool_retry_policy(tool_name)
+
+    def _effective_tool_retry_policy(self, tool_name):
+        """The one RetryPolicy that governs every retry of ``tool_name``.
+
+        Precedence: per-tool ``.retry_policy`` > agent-level
+        ``ToolConfig(retry_policy=...)`` > a policy translated from the
+        ExecutionConfig spelling of the same knobs.
+
+        ``ExecutionConfig`` speaks a second vocabulary for the same thing --
+        ``max_retry_limit`` counts *retries* (so total runs = limit + 1),
+        ``retry_initial_delay`` is in *seconds*, and ``retry_jitter`` is a
+        *fraction*. Translating it here, once, is what keeps the two retry loops
+        from running two independent budgets in two different units.
+        """
         from ..tools.retry import RetryPolicy
-        
+
         # Check for tool-level retry policy first
         tools = getattr(self, 'tools', [])
         # Handle non-iterable tools (e.g., single MCP instance)
@@ -2927,16 +3019,43 @@ class ToolExecutionMixin:
             tool_policy = getattr(tool, 'retry_policy', None) if hasattr(tool, 'retry_policy') else None
             if tool_name_attr == tool_name and tool_policy is not None:
                 return tool_policy
-        
+
         # Check for agent-level retry policy
         agent_policy = getattr(self, '_tool_retry_policy', None)
         if agent_policy is not None:
             return agent_policy
-        
-        # Return default retry policy (cached class-level instance)
-        if not hasattr(ToolExecutionMixin, '_default_retry_policy'):
-            ToolExecutionMixin._default_retry_policy = RetryPolicy()
-        return ToolExecutionMixin._default_retry_policy
+
+        # Translate the ExecutionConfig spelling into the RetryPolicy vocabulary.
+        execution_config = getattr(self, '_execution_config', None)
+        if execution_config is None:
+            max_retry_limit = getattr(self, 'max_retry_limit', 2)
+            initial_delay_s, backoff_factor, jitter_fraction = 1.0, 2.0, 0.1
+        else:
+            max_retry_limit = execution_config.max_retry_limit
+            initial_delay_s = execution_config.retry_initial_delay
+            backoff_factor = execution_config.retry_backoff_factor
+            jitter_fraction = execution_config.retry_jitter
+
+        cached = getattr(self, '_translated_retry_policy', None)
+        key = (max_retry_limit, initial_delay_s, backoff_factor, jitter_fraction)
+        if cached is not None and cached[0] == key:
+            return cached[1]
+        # Reproduce BackoffPolicy.delay's 60s base cap so ExecutionConfig users
+        # keep the exact same delay ceiling they had before this consolidation
+        # (RetryPolicy defaults to a 30s cap). initial_delay_ms must not exceed
+        # max_delay_ms per RetryPolicy validation, so lift the cap to whichever
+        # is larger.
+        initial_delay_ms = int(initial_delay_s * 1000)
+        policy = RetryPolicy(
+            max_attempts=max(1, max_retry_limit + 1),
+            backoff_factor=max(1.0, backoff_factor),
+            initial_delay_ms=initial_delay_ms,
+            max_delay_ms=max(60000, initial_delay_ms),
+            jitter=jitter_fraction > 0,
+            jitter_factor=min(max(jitter_fraction, 0.0), 1.0),
+        )
+        self._translated_retry_policy = (key, policy)
+        return policy
 
     def _classify_error_type(self, error_dict, exception):
         """Classify error type for retry policy matching.

src/praisonai-agents/praisonaiagents/agent/tools_placement.py

@@ -0,0 +1,129 @@
+"""Give a single Agent's tools a sandbox, without touching its thinking.
+
+``AgentFlow``/``AgentTeam`` already share one sandbox across a run and tear it
+down when the run ends. A lone ``Agent(tools_run_on=...)`` has no "run" to hang
+that lifetime on -- it is just an object someone calls ``chat()`` on, possibly
+many times -- so the sandbox is created on first use and lives as long as the
+agent does. That is deliberate: a file written in turn 1 is still there in turn
+2, which is what anyone who asked for a working directory expects.
+
+Attachment is idempotent and happens at the top of ``chat``/``achat``/
+``_start_stream``, and in ``AgentTeam`` before the task's tool list is
+snapshotted. Doing it once, rather than wrapping each call in a context
+manager, sidesteps two real hazards in this codebase:
+
+* ``_start_stream`` is a **generator**; a context manager around the call would
+  exit before the caller consumed a single token, tearing down the sandbox while
+  the model was still using it.
+* ``_start_stream`` falls back to ``self.chat``, so a wrapper would **re-enter**
+  itself and provision a second sandbox.
+
+Lifetime is tied to the agent by ``weakref.finalize`` rather than a process-exit
+hook over a weak set. The weak set leaked: an agent and its ``SharedCompute``
+reference each other, so a dropped agent was reclaimed by the cycle collector,
+vanished from the set, and its sandbox was never shut down. Gateways clone an
+agent **per request**, so that was one orphaned container per request.
+"""
+
+from __future__ import annotations
+
+import logging
+import threading
+import weakref
+from typing import Any
+
+logger = logging.getLogger(__name__)
+
+# One lock per agent, so two unrelated agents never wait on each other and no
+# lock is ever held while another agent's provider is being imported. Held only
+# by ensure_tools_placed/close_tools_sandbox for that one agent.
+_AGENT_LOCKS: "weakref.WeakKeyDictionary" = weakref.WeakKeyDictionary()
+_LOCKS_GUARD = threading.Lock()
+
+
+def _lock_for(agent: Any) -> threading.Lock:
+    """The placement lock for ``agent``, created on demand.
+
+    ``_LOCKS_GUARD`` covers only this dictionary lookup -- microseconds -- never
+    provider construction or attachment. An earlier version held one global lock
+    across ``SharedCompute(...)``, which imports the provider SDK: unrelated
+    agents serialised behind each other, and a slow import could stall
+    interpreter exit.
+    """
+    with _LOCKS_GUARD:
+        lock = _AGENT_LOCKS.get(agent)
+        if lock is None:
+            lock = threading.Lock()
+            try:
+                _AGENT_LOCKS[agent] = lock
+            except TypeError:  # pragma: no cover - unhashable/non-weakrefable
+                pass
+        return lock
+
+
+def _shutdown_orphan(shared: Any, name: str) -> None:
+    """Tear down a sandbox whose agent is gone.
+
+    Registered with ``weakref.finalize``, so it runs when the agent is collected
+    *and* at interpreter exit. It must not close over the agent -- that would
+    keep it alive and defeat the whole point.
+    """
+    try:
+        shared.shutdown()
+        logger.debug("[tools_placement] released the sandbox belonging to %s", name)
+    except Exception as exc:  # pragma: no cover - teardown must stay quiet
+        # Loud enough to notice: a failed release means a container, or a
+        # billed cloud instance, is still running after the agent is gone.
+        logger.warning(
+            "[tools_placement] could not release the sandbox for %s: %s", name, exc
+        )
+
+
+def ensure_tools_placed(agent: Any) -> None:
+    """Attach the sandbox tools to ``agent`` if it asked for them. Idempotent.
+
+    Called at the top of every entry point, so it must be cheap when there is
+    nothing to do -- which is the overwhelmingly common case.
+    """
+    place = getattr(agent, "tools_run_on", None)
+    if place is None or getattr(agent, "_tools_sandbox", None) is not None:
+        return
+
+    from ..managed.shared_compute import SharedCompute
+
+    with _lock_for(agent):
+        # Re-check under the lock: two threads calling chat() at once must not
+        # each provision a sandbox and leak one.
+        if getattr(agent, "_tools_sandbox", None) is not None:
+            return
+
+        shared = SharedCompute(place)
+        # override_declared: this sandbox exists *because* the agent named a
+        # place, so the skip-if-declared rule must not skip the agent itself.
+        shared.attach([agent], override_declared=True)
+        agent._tools_sandbox = shared
+        agent._tools_sandbox_finalizer = weakref.finalize(
+            agent, _shutdown_orphan, shared, str(getattr(agent, "name", agent))
+        )
+
+    logger.info("[tools_placement] %r tools now run on %s", getattr(agent, "name", agent), place)
+
+
+def close_tools_sandbox(agent: Any) -> None:
+    """Tear down the agent's sandbox and put its original tools back."""
+    with _lock_for(agent):
+        # Everything happens under the lock. Clearing the marker first and
+        # shutting down afterwards left a window where a concurrent chat()
+        # re-attached and the in-flight shutdown then restored over it, leaving
+        # the agent marked as sandboxed while holding no sandbox tools -- a
+        # state ensure_tools_placed() could never repair, because its idempotent
+        # early return sees the marker and does nothing.
+        shared = getattr(agent, "_tools_sandbox", None)
+        if shared is None:
+            return
+        finalizer = getattr(agent, "_tools_sandbox_finalizer", None)
+        if finalizer is not None:
+            finalizer.detach()          # we are shutting down explicitly
+            agent._tools_sandbox_finalizer = None
+        shared.shutdown()
+        agent._tools_sandbox = None

src/praisonai-agents/praisonaiagents/agent/unified_execution_mixin.py

@@ -282,7 +282,7 @@ class UnifiedExecutionMixin:
                     task_id=task_id,
                     config=config,
                     tool_choice=tool_choice,
-                    parallel_tool_calls=getattr(self, 'parallel_tool_calls', False)  # Gap 2 integration
+                    parallel_tool_calls=getattr(getattr(self, "execution", None), "parallel_tool_calls", False)  # Gap 2 integration
                 )
             else:
                 # Standard LiteLLM path - delegate to existing LLM class
@@ -304,7 +304,7 @@ class UnifiedExecutionMixin:
                     task_id=task_id,
                     config=config,
                     tool_choice=tool_choice,
-                    parallel_tool_calls=getattr(self, 'parallel_tool_calls', False)  # Gap 2 integration
+                    parallel_tool_calls=getattr(getattr(self, "execution", None), "parallel_tool_calls", False)  # Gap 2 integration
                 )
             
             # Store response in memory if enabled

src/praisonai-agents/praisonaiagents/agent/video_agent.py

@@ -158,9 +158,8 @@ class VideoAgent:
             verbose: Verbosity level for output
             output: Output preset ("silent", "verbose", etc.)
         """
-        # Handle model= alias for llm=
-        if llm is None and model is not None:
-            llm = model
+        from ..utils.model_alias import resolve_model_name
+        llm = resolve_model_name(llm, model, "VideoAgent")
         
         # Store core identity
         self.name = name or "VideoAgent"

src/praisonai-agents/praisonaiagents/agent/vision_agent.py

@@ -112,8 +112,8 @@ class VisionAgent:
                 - VisionConfig: Full configuration object
             verbose: Verbosity level for output
         """
-        if llm is None and model is not None:
-            llm = model
+        from ..utils.model_alias import resolve_model_name
+        llm = resolve_model_name(llm, model, "VisionAgent")
         
         self.name = name or "VisionAgent"
         self.instructions = instructions

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

@@ -2,6 +2,8 @@ import os
 import time
 import json
 import logging
+import contextlib
+import contextvars
 import threading
 from praisonaiagents._logging import get_logger
 from typing import Any, Dict, Optional, List, Tuple, Callable
@@ -41,18 +43,13 @@ class TaskStatus(Enum):
 logger = get_logger(__name__)
 
 
-def _launch_auth_token() -> Optional[str]:
-    return os.environ.get("PRAISONAI_LAUNCH_AUTH_TOKEN")
-
-
-def _authorise_launch_request(request) -> bool:
-    token = _launch_auth_token()
-    if not token:
-        return True
-    auth = request.headers.get("Authorization", "")
-    if auth.startswith("Bearer ") and auth[7:] == token:
-        return True
-    return request.headers.get("X-Auth-Token") == token
+# Request authorisation + fail-closed bind guard are shared with Agent.launch
+# so the single-agent and multi-agent serving paths cannot drift apart.
+from ..agent.launch_security import (
+    launch_auth_token as _launch_auth_token,
+    authorise_launch_request as _authorise_launch_request,
+    resolve_launch_host as _resolve_launch_host,
+)
 
 
 # Agent server registry for thread-safe server management
@@ -375,6 +372,18 @@ def _build_execution_context(agents_instance, task_id, skip_memory_init=False):
     if llm and hasattr(llm, 'set_current_agent'):
         llm.set_current_agent(executor_agent.display_name)
 
+    # Place the agent's tools BEFORE snapshotting them. This snapshot is passed
+    # to chat() as an explicit `tools=`, which wins over `agent.tools` -- so if
+    # the sandbox attaches later (inside chat()), the model is handed the
+    # pre-attach list and every tool runs on the host while a sandbox sits
+    # provisioned and unused. Silent, and worse than it sounds: in
+    # start_for_each() row 1 got local tools and row 2 got the bridged set,
+    # because row 1's chat() mutated agent.tools in time for row 2's snapshot.
+    if executor_agent is not None and getattr(executor_agent, "tools_run_on", None) is not None:
+        from ..agent.tools_placement import ensure_tools_placed
+
+        ensure_tools_placed(executor_agent)
+
     # Ensure tools are available from both task and agent (create copy to avoid mutation)
     tools = list(task.tools or [])
     if executor_agent and executor_agent.tools:
@@ -644,7 +653,8 @@ class AgentTeam(SpawnAnnounceProtocol):
         manager_llm=None,
         name: Optional[str] = None,
         variables: Optional[Dict[str, Any]] = None,
-        llm: Optional[str] = None,  # Default LLM for all agents (API consistency)
+        llm: Optional[str] = None,  # Deprecated alias for model=
+        model: Optional[str] = None,  # Default model for members that did not name one
         # Consolidated feature params (agent-centric API)
         memory: Optional[Any] = False,  # Union[bool, MultiAgentMemoryConfig]
         planning: Optional[Any] = False,  # Union[bool, MultiAgentPlanningConfig]
@@ -660,13 +670,15 @@ class AgentTeam(SpawnAnnounceProtocol):
         reflection: Optional[Any] = None,  # Union[bool, ReflectionConfig] - self-reflection
         caching: Optional[Any] = None,  # Union[bool, CachingConfig] - caching
         learn: Optional[Any] = None,  # Union[bool, LearnConfig] - continuous learning
-        # Union[str, ComputeProviderProtocol] - run every agent's shell/file
-        # tools in ONE shared remote sandbox ("docker", "e2b", "modal",
-        # "daytona", "flyio", "tenki", "local"). Agents share /workspace, so
-        # files written by one agent are visible to the others. Orchestration
-        # stays local. Pass a configured provider instance instead of a name to
-        # customise image/resources.
-        compute: Optional[Any] = None,
+        # Union[str, ComputeProviderProtocol] - where every agent's shell/file
+        # tools run: "docker", "e2b", "modal", "daytona", "flyio", "tenki",
+        # "local", "subprocess", "sandlock", "ssh", "novita". The team shares
+        # ONE sandbox and one working directory, so files written by one agent
+        # are visible to the others. Orchestration and thinking stay local.
+        # Pass a configured provider instance to customise resources.
+        tools_run_on: Optional[Any] = None,
+        # Accepted only to fail with a useful message -- see resolve_placement.
+        run_on: Optional[Any] = None,
     ):
         """
         Initialize AgentManager with consolidated feature parameters.
@@ -677,8 +689,11 @@ class AgentTeam(SpawnAnnounceProtocol):
             process: Execution process type ("sequential", "workflow", "hierarchical").
                 For parallel fan-out, set async_execution=True on individual Task
                 objects within a "workflow" or "sequential" process.
-            manager_llm: LLM model for manager agent
-            llm: Default LLM model for all agents
+            manager_llm: Model for the hierarchical manager agent. Unrelated to
+                model=/llm=; it never touches the members.
+            llm: Deprecated alias for model=. Passing both raises TypeError.
+            model: Default model for members that did not name one themselves.
+                An agent constructed with its own llm=/model=/auth= keeps it.
             name: Name for this agent collection
             variables: Global variables for substitution
             memory: Memory configuration (bool | MultiAgentMemoryConfig)
@@ -696,13 +711,37 @@ class AgentTeam(SpawnAnnounceProtocol):
         """
         # Store new params for propagation to agents
         self._learn = learn
-        self.llm = llm  # Store default LLM for API consistency
+        # model= is the canonical name; llm= is the deprecated alias. Passing
+        # both is refused rather than silently resolved (see resolve_model_alias).
+        # The resolved value is the default model for the members: applied to
+        # every agent that never named a model of its own; an agent that did
+        # keeps it. Same rule AgentFlow uses for the agents it builds.
+        from ..utils.model_alias import resolve_model_alias
+        self.llm = resolve_model_alias(llm, model, type(self).__name__)
+        if self.llm:
+            for _member in (agents.values() if isinstance(agents, dict) else (agents or [])):
+                _apply = getattr(_member, "_apply_default_llm", None)
+                if callable(_apply):
+                    _apply(self.llm)
         self._autonomy = autonomy
         self._knowledge = knowledge
         self._guardrails = guardrails
         self._web = web
         self._reflection = reflection
         self._caching = caching
+        # Warn about params that are accepted for API symmetry with Agent but are
+        # not yet applied at the team level, so a truthy value stops being a silent
+        # no-op. Pass these to individual Agent(...) instances instead.
+        _unwired = {
+            "guardrails": guardrails, "web": web, "reflection": reflection,
+            "caching": caching, "learn": learn, "knowledge": knowledge,
+        }
+        _set = [k for k, v in _unwired.items() if v]
+        if _set:
+            logging.warning(
+                f"AgentTeam received {_set} but does not yet apply them at the team "
+                "level; pass these to individual Agent(...) instances instead."
+            )
         # ─────────────────────────────────────────────────────────────────────
         # Extract values from consolidated params using UNIFIED CANONICAL resolver
         # Precedence: Instance > Config > Dict > Array > String > Bool > Default
@@ -898,7 +937,19 @@ class AgentTeam(SpawnAnnounceProtocol):
         self.stream = _stream
         self.name = name
         # Remote execution: one shared sandbox for every agent on this team.
-        self.compute = compute
+        from ..agent.placement import resolve_placement
+
+        resolve_placement(
+            "AgentTeam", run_on=run_on, tools_run_on=tools_run_on,
+            supports_run_on=False,
+        )
+        self.tools_run_on = tools_run_on
+        # Kept readable rather than removed. `team.run_on` used to exist, and
+        # code doing getattr(team, "run_on", None) for introspection or
+        # serialisation would otherwise start raising AttributeError. None is
+        # the honest value: a team orchestrates locally, so nothing about it
+        # runs wholly on a managed runtime.
+        self.run_on = None
 
         # Callbacks for workflow execution
         self.on_task_start = _on_task_start
@@ -1425,6 +1476,73 @@ class AgentTeam(SpawnAnnounceProtocol):
                 raise result
         return results
 
+    def _depends_on_pending(self, task, pending):
+        """True if ``task`` depends on a task still pending in the batch.
+
+        ``pending`` is a list of ``(task_id, coroutine)`` pairs already queued
+        for parallel execution. If one of the task's dependencies is in that
+        set, its result isn't available yet, so the batch must be flushed before
+        queuing this task (otherwise the dependent builds its prompt from an
+        empty result for the still-in-progress dependency).
+
+        Two dependency edges are checked, since ``_build_task_context`` reads
+        from both:
+          - ``task.context``: explicit context tasks (by object ``id``).
+          - ``task.previous_tasks``: workflow ``next_tasks`` edges, stored as
+            task *names*; without this a workflow successor could be queued in
+            the same async batch as its predecessor and read an empty result.
+        """
+        if not pending:
+            return False
+        pending_ids = {tid for tid, _ in pending}
+
+        deps = getattr(task, 'context', None) or []
+        if any(getattr(dep, 'id', None) in pending_ids for dep in deps):
+            return True
+
+        previous = getattr(task, 'previous_tasks', None) or []
+        if previous:
+            pending_names = {
+                self.tasks[tid].name
+                for tid in pending_ids
+                if tid in self.tasks
+            }
+            if any(name in pending_names for name in previous):
+                return True
+
+        return False
+
+    def _deps_failed(self, task_id):
+        """True if any dependency of ``task_id`` finished in a failed state.
+
+        asequential()/aworkflow() run their own failure check inside the
+        generator, but async tasks are buffered and drained later, so an
+        upstream async task may not have failed yet when the generator yielded
+        the dependent. Re-checking here — after pending async tasks are flushed
+        — makes the failure cascade fire for the async_execution path too, for
+        both dependency edges ``_build_task_context`` reads (context tasks and
+        workflow ``previous_tasks``).
+        """
+        task = self.tasks.get(task_id)
+        if task is None:
+            return False
+
+        for dep in getattr(task, 'context', None) or []:
+            dep_id = getattr(dep, 'id', None)
+            if dep_id in self.tasks and self.tasks[dep_id].status == "failed":
+                return True
+
+        previous = getattr(task, 'previous_tasks', None) or []
+        if previous:
+            by_name = {t.name: t for t in self.tasks.values()}
+            if any(
+                name in by_name and by_name[name].status == "failed"
+                for name in previous
+            ):
+                return True
+
+        return False
+
     async def arun_all_tasks(self):
         """Async version of run_all_tasks method"""
         process = Process(
@@ -1436,14 +1554,26 @@ class AgentTeam(SpawnAnnounceProtocol):
         )
         
         if self.process == "workflow":
-            tasks_to_run = []
+            tasks_to_run = []  # list of (task_id, coroutine)
             async for task_id in process.aworkflow():
-                if self.tasks[task_id].async_execution:
-                    tasks_to_run.append(self.arun_task(task_id))
+                task = self.tasks[task_id]
+                if task.async_execution:
+                    # If this async task depends on another async task still
+                    # pending in the current batch, flush first so its context
+                    # is available before this one reads it.
+                    if self._depends_on_pending(task, tasks_to_run):
+                        await self._gather_with_isolation([c for _, c in tasks_to_run])
+                        tasks_to_run = []
+                        # A just-flushed prerequisite may now be failed; don't
+                        # queue a dependent that would run with missing context.
+                        if self._deps_failed(task_id):
+                            self.tasks[task_id].status = "failed"
+                            continue
+                    tasks_to_run.append((task_id, self.arun_task(task_id)))
                 else:
                     # If we encounter a sync task, we must wait for the previous async tasks to finish.
                     if tasks_to_run:
-                        await self._gather_with_isolation(tasks_to_run)
+                        await self._gather_with_isolation([c for _, c in tasks_to_run])
                         tasks_to_run = []
                     
                     # Run sync task in an executor to avoid blocking the event loop
@@ -1453,25 +1583,42 @@ class AgentTeam(SpawnAnnounceProtocol):
                     await loop.run_in_executor(None, copy_context_to_callable(lambda tid=task_id: self.run_task(tid)))
 
             if tasks_to_run:
-                await self._gather_with_isolation(tasks_to_run)
+                await self._gather_with_isolation([c for _, c in tasks_to_run])
                 
         elif self.process == "sequential":
-            async_tasks_to_run = []
+            async_tasks_to_run = []  # list of (task_id, coroutine)
             
             async def flush_async_tasks():
                 """Execute all pending async tasks"""
                 nonlocal async_tasks_to_run
                 if async_tasks_to_run:
-                    await self._gather_with_isolation(async_tasks_to_run)
+                    await self._gather_with_isolation([c for _, c in async_tasks_to_run])
                     async_tasks_to_run = []
-            
+
             async for task_id in process.asequential():
-                if self.tasks[task_id].async_execution:
+                task = self.tasks[task_id]
+                if task.async_execution:
+                    # If this async task depends on another async task still
+                    # pending in the current batch, flush first so its context
+                    # is available before this one reads it (avoids a same-batch
+                    # race that silently substitutes empty dependency context).
+                    if self._depends_on_pending(task, async_tasks_to_run):
+                        await flush_async_tasks()
+                        # A just-flushed prerequisite may now be failed; don't
+                        # queue a dependent that would run with missing context.
+                        if self._deps_failed(task_id):
+                            self.tasks[task_id].status = "failed"
+                            continue
                     # Collect async tasks to run in parallel
-                    async_tasks_to_run.append(self.arun_task(task_id))
+                    async_tasks_to_run.append((task_id, self.arun_task(task_id)))
                 else:
                     # Before running a sync task, execute all pending async tasks
                     await flush_async_tasks()
+                    # A just-flushed async prerequisite may now be failed; skip
+                    # the dependent so we don't run it with missing upstream context.
+                    if self._deps_failed(task_id):
+                        self.tasks[task_id].status = "failed"
+                        continue
                     # Run sync task in an executor to avoid blocking the event loop
                     # Use copy_context_to_callable to propagate contextvars (needed for trace emission)
                     from ..trace.context_events import copy_context_to_callable
@@ -1509,6 +1656,62 @@ class AgentTeam(SpawnAnnounceProtocol):
             except StopAsyncIteration:
                 pass
 
+
+    # ── shared sandbox lifetime ──────────────────────────────────────────────
+    # A ContextVar, not an instance attribute: the previous version set
+    # `self.run_on = None` for the duration of the run, so a second caller
+    # inspecting the team mid-run saw "no sandbox" and any concurrent start()
+    # executed every tool on the host. Depth lives with the caller, not on the
+    # shared object, so the team's declared placement is always readable.
+    #
+    # The depth is keyed PER TEAM, not shared class-wide. A single integer here
+    # meant any team started while another team's run was on the stack saw
+    # depth==1 and skipped its own sandbox entirely -- so a nested team ran on
+    # the host, and a team sharing an agent with the outer run executed on the
+    # OUTER team's sandbox.
+    #
+    # ContextVar, not threading.local: two coroutines that `astart()` the SAME
+    # team under one asyncio.gather run on one thread, so thread-local state is
+    # shared between them -- the second saw the first's depth==1, skipped its
+    # own sandbox, and ran against a sandbox the first could tear down mid-flight.
+    # A ContextVar is copied per task at gather/create_task time, so each run
+    # gets its own depth map; it is also per-thread, so sync callers stay
+    # isolated too. The value is an immutable frozen mapping replaced via .set()
+    # (never mutated in place), because a copied context shares the *reference*
+    # to a mutable dict and in-place mutation would leak straight back across
+    # the tasks the copy was meant to separate.
+    _tools_scope_depths: "contextvars.ContextVar" = contextvars.ContextVar(
+        "praisonai_tools_scope_depths", default=()
+    )
+
+    def _needs_tools_scope(self) -> bool:
+        """True when this call should provision the team's shared sandbox."""
+        if getattr(self, "tools_run_on", None) is None:
+            return False
+        return dict(self._tools_scope_depths.get()).get(id(self), 0) == 0
+
+    @contextlib.contextmanager
+    def _shared_tools_scope(self):
+        """Hold one sandbox for everything nested inside.
+
+        Re-entrant by design: ``start_for_each`` opens the scope once for the
+        whole batch, so the ``start()`` call it makes per input row finds the
+        scope already open and reuses that sandbox instead of provisioning and
+        destroying one per row.
+        """
+        from ..managed.shared_compute import SharedCompute
+
+        key = id(self)
+        depths = dict(self._tools_scope_depths.get())
+        depths[key] = depths.get(key, 0) + 1
+        token = self._tools_scope_depths.set(tuple(sorted(depths.items())))
+        try:
+            with SharedCompute(self.tools_run_on) as shared:
+                shared.attach(list(self.agents or []))
+                yield shared
+        finally:
+            self._tools_scope_depths.reset(token)
+
     async def astart(self, content=None, return_dict=False, **kwargs):
         """Async version of start method.
         
@@ -1517,6 +1720,12 @@ class AgentTeam(SpawnAnnounceProtocol):
             return_dict: If True, returns the full results dictionary instead of only the final response
             **kwargs: Additional arguments
         """
+        # Same shared sandbox as start(). Without this, an async team with
+        # tools_run_on= ran every tool on the host and said nothing.
+        if self._needs_tools_scope():
+            with self._shared_tools_scope():
+                return await self.astart(content=content, return_dict=return_dict, **kwargs)
+
         # Track execution via telemetry
         if hasattr(self, '_telemetry') and self._telemetry:
             self._telemetry.track_agent_execution(self.name, success=True, async_mode=True)
@@ -1737,6 +1946,21 @@ class AgentTeam(SpawnAnnounceProtocol):
             return str(agent[0])
         return None
 
+    def __repr__(self):
+        """Show where this team's work happens, not just its name."""
+        from ..agent.execution_location import repr_fields
+
+        return (
+            f"AgentTeam(name={self.name!r}, agents={len(self.agents or [])}, "
+            f"{repr_fields(self, shared=True)})"
+        )
+
+    def where_does_it_run(self) -> str:
+        """Plain-English answer to "where does my code actually run?"."""
+        from ..agent.execution_location import explain
+
+        return explain(self, shared=True)
+
     def start(self, content=None, return_dict=False, output=None, **kwargs):
         """Start agent execution with verbose output (beginner-friendly).
         
@@ -1762,19 +1986,11 @@ class AgentTeam(SpawnAnnounceProtocol):
             ```
         """
         # Remote execution: provision ONE sandbox shared by every agent on the
-        # team, and tear it down even if execution raises. Re-enters start()
-        # with compute cleared so the wrapper applies exactly once.
-        if getattr(self, "compute", None) is not None:
-            from ..managed.shared_compute import SharedCompute
-
-            compute, self.compute = self.compute, None
-            try:
-                with SharedCompute(compute) as shared:
-                    shared.attach(list(self.agents or []))
-                    return self.start(content=content, return_dict=return_dict,
-                                      output=output, **kwargs)
-            finally:
-                self.compute = compute
+        # team, and tear it down even if execution raises.
+        if self._needs_tools_scope():
+            with self._shared_tools_scope():
+                return self.start(content=content, return_dict=return_dict,
+                                  output=output, **kwargs)
 
         # Track execution via telemetry
         if hasattr(self, '_telemetry') and self._telemetry:
@@ -2090,6 +2306,15 @@ class AgentTeam(SpawnAnnounceProtocol):
         if on_error not in ("continue", "fail_fast"):
             raise ValueError("on_error must be 'continue' or 'fail_fast'")
 
+        # One sandbox for the whole batch. Without this the per-row start()
+        # call provisions and destroys its own, so a 50-row batch paid for 50
+        # sandboxes and no row could see what an earlier row wrote.
+        if self._needs_tools_scope():
+            with self._shared_tools_scope():
+                return self.start_for_each(
+                    inputs, on_error=on_error, output=output, **kwargs
+                )
+
         batch_id = f"batch_{uuid.uuid4().hex[:12]}"
         saved_variables = self.variables
         task_snapshot = self._snapshot_task_state()
@@ -2126,6 +2351,13 @@ class AgentTeam(SpawnAnnounceProtocol):
         if on_error not in ("continue", "fail_fast"):
             raise ValueError("on_error must be 'continue' or 'fail_fast'")
 
+        # One sandbox for the whole batch -- see start_for_each.
+        if self._needs_tools_scope():
+            with self._shared_tools_scope():
+                return await self.astart_for_each(
+                    inputs, on_error=on_error, **kwargs
+                )
+
         batch_id = f"batch_{uuid.uuid4().hex[:12]}"
         saved_variables = self.variables
         task_snapshot = self._snapshot_task_state()
@@ -2407,7 +2639,7 @@ class AgentTeam(SpawnAnnounceProtocol):
         
         print("="*50 + "\n")
         
-    def launch(self, path: str = '/agents', port: int = 8000, host: str = '0.0.0.0', debug: bool = False, protocol: str = "http"):
+    def launch(self, path: str = '/agents', port: int = 8000, host: str = '127.0.0.1', debug: bool = False, protocol: str = "http"):
         """
         Launch all agents as a single API endpoint (HTTP) or an MCP server. 
         In HTTP mode, the endpoint accepts a query and processes it through all agents in sequence.
@@ -2418,7 +2650,11 @@ class AgentTeam(SpawnAnnounceProtocol):
         Args:
             path: API endpoint path (default: '/agents') for HTTP. Ignored in MCP mode.
             port: Server port (default: 8000)
-            host: Server host (default: '0.0.0.0')
+            host: Server host (default: '127.0.0.1'; changed from '0.0.0.0' to
+                avoid binding all interfaces by default — pass '0.0.0.0'
+                explicitly to expose externally). Binding a non-loopback host
+                without ``PRAISONAI_LAUNCH_AUTH_TOKEN`` set auto-generates and
+                prints a one-time bearer token.
             debug: Enable debug mode for uvicorn (default: False)
             protocol: "http" to launch as FastAPI, "mcp" to serve over MCP via
                 ``praisonai-mcp`` (install with ``pip install praisonai-mcp``).
@@ -2428,7 +2664,7 @@ class AgentTeam(SpawnAnnounceProtocol):
         """
         if protocol == "http":
             # Use centralized server registry
-            
+
             if not self.agents:
                 logging.warning("No agents to launch for HTTP mode. Add agents to the Agents instance first.")
                 return
@@ -2457,6 +2693,13 @@ class AgentTeam(SpawnAnnounceProtocol):
                 print("\nOr install all API dependencies with:")
                 print("pip install 'praisonaiagents[api]'")
                 return None
+
+            # Fail-closed bind guard shared with Agent.launch: never serve
+            # keyless on a non-loopback host. Applied only after the no-agents
+            # and dependency guards above, so a launch that bails out cannot
+            # mutate the process-wide launch token (which every route reads per
+            # request) and retroactively 401 an already-running endpoint.
+            host = _resolve_launch_host(host)
             
             # Thread-safe initialization of FastAPI app
             app, is_new = _server_registry.get_or_create_app(
@@ -3095,9 +3338,14 @@ class AgentTeam(SpawnAnnounceProtocol):
         console.print(f"[dim]Progress: [{'█' * 30}] 100%[/dim]")
         console.print(f"[green]Completed {completed_count}/{len(self.tasks)} tasks![/green]\n")
         
-        # Restore original tasks reference for result retrieval
-        self._plan_tasks = self.tasks.copy()
-        # Keep plan tasks for results but note original tasks are preserved in _plan_tasks
+        # Keep plan-step tasks/results under _plan_tasks for introspection,
+        # then restore the user's original task set as the canonical self.tasks.
+        self._plan_tasks = self.tasks
+        self.tasks = original_tasks
+        with self._task_id_lock:
+            self.task_id_counter = (
+                max(original_tasks.keys()) + 1 if original_tasks else 0
+            )
 
     # Resource Lifecycle Management
     def close(self) -> None:

src/praisonai-agents/praisonaiagents/approval/backends.py

@@ -81,12 +81,14 @@ class ConsoleBackend:
     """Interactive Rich terminal prompt.  Default for CLI usage."""
 
     def _prompt_user(self, request: ApprovalRequest):
-        """Show Rich panel and ask once/session/always/no.
+        """Show Rich panel and ask once/session/always/no/deny-with-guidance.
 
-        Returns a ``(approved, scope, scope_pattern)`` tuple where ``scope`` is
-        one of ``"once"`` / ``"session"`` / ``"always"`` and ``scope_pattern``
-        is the reusable target to persist for ``always`` (``None`` otherwise).
-        A denial returns ``(False, "once", None)``.
+        Returns a ``(approved, scope, scope_pattern, feedback)`` tuple where
+        ``scope`` is one of ``"once"`` / ``"session"`` / ``"always"`` and
+        ``scope_pattern`` is the reusable target to persist for ``always``
+        (``None`` otherwise). ``feedback`` carries optional free-text guidance
+        captured by the "deny & redirect" choice (``None`` otherwise).
+        A plain denial returns ``(False, "once", None, None)``.
         """
         Console = _get_rich_console()
         Panel = _get_rich_panel()
@@ -126,28 +128,35 @@ class ConsoleBackend:
         try:
             console.print(
                 f"[{risk_color}]Allow {request.tool_name}?[/{risk_color}]  "
-                f"[o] once   [s] this session   [a] {always_label}   [n] no"
+                f"[o] once   [s] this session   [a] {always_label}   "
+                f"[n] no   [d] deny & redirect"
             )
             choice = Prompt.ask(
                 "Choice",
-                choices=["o", "s", "a", "n"],
+                choices=["o", "s", "a", "n", "d"],
                 default="n",
             )
         except (KeyboardInterrupt, EOFError):
-            return (False, "once", None)
+            return (False, "once", None, None)
 
         if choice == "n":
-            return (False, "once", None)
+            return (False, "once", None, None)
+        if choice == "d":
+            try:
+                feedback = Prompt.ask("What should the agent do instead?").strip()
+            except (KeyboardInterrupt, EOFError):
+                feedback = ""
+            return (False, "once", None, feedback or None)
         if choice == "s":
-            return (True, "session", None)
+            return (True, "session", None, None)
         if choice == "a":
-            return (True, "always", suggested or None)
-        return (True, "once", None)
+            return (True, "always", suggested or None, None)
+        return (True, "once", None, None)
 
     def request_approval_sync(self, request: ApprovalRequest) -> ApprovalDecision:
         """Synchronous approval via Rich console prompt."""
         try:
-            approved, scope, scope_pattern = self._prompt_user(request)
+            approved, scope, scope_pattern, feedback = self._prompt_user(request)
             if approved:
                 return ApprovalDecision(
                     approved=True,
@@ -156,7 +165,13 @@ class ConsoleBackend:
                     scope=scope,
                     scope_pattern=scope_pattern,
                 )
-            return ApprovalDecision(approved=False, reason="User denied", approver="console")
+            reason = f"User denied: {feedback}" if feedback else "User denied"
+            return ApprovalDecision(
+                approved=False,
+                reason=reason,
+                approver="console",
+                feedback=feedback,
+            )
         except Exception as e:
             logger.error("Console approval error: %s", e)
             return ApprovalDecision(approved=False, reason=f"Approval error: {e}")

src/praisonai-agents/praisonaiagents/approval/protocols.py

@@ -65,6 +65,12 @@ class ApprovalDecision:
                        ``session``/``always`` scopes (e.g. ``"bash:git status *"``
                        or ``"edit:src/app.py"``). When ``None`` the registry
                        derives one from the tool call.
+        feedback:      Optional free-text guidance supplied by the user when
+                       denying a tool call ("deny with guidance"). When present
+                       on a denial it is surfaced back to the model as a
+                       course-correction for the next turn instead of a bare
+                       "denied" result. ``None`` (default) keeps today's
+                       terminal-deny semantics.
     """
 
     approved: bool
@@ -74,6 +80,7 @@ class ApprovalDecision:
     metadata: Dict[str, Any] = field(default_factory=dict)
     scope: str = "once"
     scope_pattern: Optional[str] = None
+    feedback: Optional[str] = None
 
 
 @dataclass

src/praisonai-agents/praisonaiagents/auth/__init__.py

@@ -2,7 +2,10 @@
 Authentication & Authorization Protocols for PraisonAI Agents.
 
 This package provides protocol-driven auth interfaces for the core SDK.
-Implementations (JWT, OAuth, etc.) live in the platform package.
+Implementations (JWT, OAuth, etc.) live outside it — implement the protocols
+against your own store, or use the reference implementation in the separate
+``praisonai-platform`` package
+(https://github.com/MervinPraison/PraisonAI-Platform).
 
 Quick start::
 

src/praisonai-agents/praisonaiagents/auth/protocols.py

@@ -155,11 +155,16 @@ class WorkspaceContextProtocol(Protocol):
     """Protocol for providing workspace context to agents.
 
     Enables agents to access workspace-specific settings, instructions,
-    and agent configurations from the platform layer.
+    and agent configurations from a platform layer.
+
+    Implementations live outside this package. The reference one is
+    ``PlatformWorkspaceContext`` in the ``praisonai-platform`` package
+    (https://github.com/MervinPraison/PraisonAI-Platform); implement this
+    protocol yourself to back agent configuration with your own store.
 
     Example::
 
-        class PlatformWorkspaceContext:
+        class MyWorkspaceContext:
             async def get_workspace_context(self, workspace_id):
                 ws = await db.get_workspace(workspace_id)
                 return ws.context

src/praisonai-agents/praisonaiagents/bots/protocols.py

@@ -109,6 +109,11 @@ class PlatformCapabilities:
             descriptor only; the durable outbox does not forward a token on
             resend, so adapters relying on provider-side dedupe should also set
             ``reconciles_unknown_send`` to get effectively-once delivery.
+        supports_threads: Whether the adapter can open a new thread/topic under
+            a chat (e.g. Telegram forum topics, Discord threads, Slack thread
+            anchoring). When False, ``create_thread`` returns a typed
+            ``unsupported`` outcome and callers fall back to the parent channel
+            rather than raising.
     """
     
     max_message_length: int = 4096
@@ -126,6 +131,7 @@ class PlatformCapabilities:
     reconciles_unknown_send: bool = False
     supports_idempotency_token: bool = False
     supports_media: bool = False
+    supports_threads: bool = False
     
     def to_dict(self) -> Dict[str, Any]:
         """Convert to dictionary."""
@@ -145,6 +151,7 @@ class PlatformCapabilities:
             "reconciles_unknown_send": self.reconciles_unknown_send,
             "supports_idempotency_token": self.supports_idempotency_token,
             "supports_media": self.supports_media,
+            "supports_threads": self.supports_threads,
         }
     
     @classmethod
@@ -166,6 +173,7 @@ class PlatformCapabilities:
             reconciles_unknown_send=data.get("reconciles_unknown_send", False),
             supports_idempotency_token=data.get("supports_idempotency_token", False),
             supports_media=data.get("supports_media", False),
+            supports_threads=data.get("supports_threads", False),
         )
 
 

src/praisonai-agents/praisonaiagents/bus/__init__.py

@@ -32,6 +32,8 @@ Usage:
     bus.publish("session.created", {"session_id": "abc123"})
 """
 
+from .._lazy import create_lazy_getattr
+
 __all__ = [
     "EventBus",
     "Event",
@@ -41,31 +43,13 @@ __all__ = [
     "set_default_bus",
 ]
 
+_LAZY_IMPORTS = {
+    "EventBus": ("praisonaiagents.bus.bus", "EventBus"),
+    "Event": ("praisonaiagents.bus.event", "Event"),
+    "EventType": ("praisonaiagents.bus.event", "EventType"),
+    "Subscriber": ("praisonaiagents.bus.bus", "Subscriber"),
+    "get_default_bus": ("praisonaiagents.bus.bus", "get_default_bus"),
+    "set_default_bus": ("praisonaiagents.bus.bus", "set_default_bus"),
+}
 
-def __getattr__(name: str):
-    """Lazy load module components."""
-    if name == "EventBus":
-        from .bus import EventBus
-        return EventBus
-    
-    if name == "Event":
-        from .event import Event
-        return Event
-    
-    if name == "EventType":
-        from .event import EventType
-        return EventType
-    
-    if name == "Subscriber":
-        from .bus import Subscriber
-        return Subscriber
-    
-    if name == "get_default_bus":
-        from .bus import get_default_bus
-        return get_default_bus
-    
-    if name == "set_default_bus":
-        from .bus import set_default_bus
-        return set_default_bus
-    
-    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
+__getattr__ = create_lazy_getattr(_LAZY_IMPORTS, __name__)

src/praisonai-agents/praisonaiagents/compaction/__init__.py

@@ -37,6 +37,8 @@ __all__ = [
     "ToolResultPrunerProtocol",
     "MessageFormatterProtocol", 
     "SummaryBuilderProtocol",
+    # Default structured summary builder
+    "DefaultSummaryBuilder",
     # Read-only recap
     "build_recap",
     "MemoryFlushResult",
@@ -75,6 +77,10 @@ def __getattr__(name: str):
         from .protocols import SummaryBuilderProtocol
         return SummaryBuilderProtocol
     
+    if name == "DefaultSummaryBuilder":
+        from .summary_builder import DefaultSummaryBuilder
+        return DefaultSummaryBuilder
+
     if name == "build_recap":
         from .recap import build_recap
         return build_recap

src/praisonai-agents/praisonaiagents/compaction/compactor.py

@@ -122,7 +122,18 @@ class ContextCompactor:
         
         # Protocol implementations (defaults to None - no heavy implementations in core)
         self.tool_pruner = tool_pruner
-        self.message_formatter = message_formatter  
+        self.message_formatter = message_formatter
+        # Default the structured summary builder when the config asks for the
+        # phase-aware template but no builder was injected, so
+        # `_build_structured_summary` renders SUMMARY_TEMPLATE instead of the
+        # "Summary of N messages" stub. The builder is deterministic and
+        # dependency-free, so this stays offline-safe and zero-overhead.
+        if summary_builder is None and getattr(self.config, "structured_template", False):
+            from .summary_builder import DefaultSummaryBuilder
+
+            summary_builder = DefaultSummaryBuilder(
+                llm_summarize_fn=self.llm_summarize_fn
+            )
         self.summary_builder = summary_builder
         
         # Anti-thrashing state tracking

src/praisonai-agents/praisonaiagents/compaction/summary_builder.py

@@ -0,0 +1,172 @@
+"""Default structured summary builder for context compaction.
+
+This is the concrete :class:`~praisonaiagents.compaction.protocols.
+SummaryBuilderProtocol` implementation that renders the declared
+:data:`~praisonaiagents.compaction.config.SUMMARY_TEMPLATE` — the phase-aware
+sections (``## Active Task`` / ``## Completed Actions`` / ``## In Progress`` /
+``## Pending Questions`` / ``## Relevant Files / Paths`` / ``## Remaining
+Work``) — so a compacted context keeps its structure instead of degrading to a
+one-line ``"Summary of N messages"`` count.
+
+Design: **deterministic, dependency-free, offline-safe.** It categorises the
+transcript into the template sections from the messages themselves (roles,
+tool calls, question marks, file-path tokens) rather than making an LLM call, so
+it never blocks the hot compaction path, never adds a runtime dependency, and
+never raises into the loop. File paths and identifiers are preserved verbatim.
+
+When an optional ``llm_summarize_fn`` is supplied the builder still renders the
+same template shape; on any failure it falls back to the deterministic render,
+honouring the issue's "fall back to the plain summary on failure" rule.
+"""
+
+from __future__ import annotations
+
+import re
+from typing import Any, Dict, List, Optional
+
+from .config import SUMMARY_TEMPLATE
+
+# Match file/path-like tokens so they can be surfaced verbatim in the
+# "Relevant Files / Paths" section (absolute/relative paths and dotted module
+# names with a recognisable extension).
+_PATH_RE = re.compile(
+    r"(?:/[^\s'\"`]+|(?:[\w.-]+/)+[\w.-]+|[\w./-]+\.[A-Za-z0-9]{1,6})"
+)
+
+_MAX_ITEMS = 12
+_MAX_LINE = 200
+
+
+def _text(content: Any) -> str:
+    """Best-effort flatten of a message ``content`` field into a string."""
+    if isinstance(content, str):
+        return content
+    if isinstance(content, list):
+        parts: List[str] = []
+        for part in content:
+            if isinstance(part, dict):
+                parts.append(str(part.get("text", "")))
+            else:
+                parts.append(str(part))
+        return " ".join(p for p in parts if p)
+    return ""
+
+
+def _clip(text: str, limit: int = _MAX_LINE) -> str:
+    text = text.strip().replace("\n", " ")
+    return text if len(text) <= limit else text[: limit - 1].rstrip() + "…"
+
+
+class DefaultSummaryBuilder:
+    """Deterministic :class:`SummaryBuilderProtocol` renderer for the template.
+
+    Args:
+        llm_summarize_fn: Optional callable ``(prompt: str) -> str`` used to
+            distil section bodies. Any failure falls back to the deterministic
+            render, so callers get a valid structured summary regardless.
+    """
+
+    def __init__(self, llm_summarize_fn: Optional[Any] = None) -> None:
+        self._llm_summarize_fn = llm_summarize_fn
+
+    # -- protocol ----------------------------------------------------------
+
+    def build_structured_summary(self, messages: List[Dict[str, Any]]) -> str:
+        """Render :data:`SUMMARY_TEMPLATE` from ``messages`` deterministically.
+
+        Never raises: on any failure returns the plain one-line summary so the
+        compaction loop keeps a valid summary string (issue fallback rule).
+        """
+        try:
+            return self._render(messages)
+        except Exception:  # noqa: BLE001 - summarisation must never crash the loop
+            return f"Summary of {len(messages)} messages"
+
+    def merge_summaries(self, previous: str, current: str) -> str:
+        """Merge a prior structured summary with the current one.
+
+        Keeps both under a single "previous / latest" framing so an iterative
+        re-compaction never discards the earlier phase structure.
+        """
+        previous = (previous or "").strip()
+        current = (current or "").strip()
+        if not previous:
+            return current
+        if not current:
+            return previous
+        return f"{previous}\n\n[Updated with newer activity]\n{current}"
+
+    # -- rendering ---------------------------------------------------------
+
+    def _render(self, messages: List[Dict[str, Any]]) -> str:
+        completed: List[str] = []
+        in_progress: List[str] = []
+        pending: List[str] = []
+        files: List[str] = []
+        active_task = ""
+
+        for msg in messages:
+            role = msg.get("role", "")
+            body = _text(msg.get("content"))
+
+            # First substantive user turn frames the active task.
+            if role == "user" and body.strip() and not active_task:
+                active_task = _clip(body)
+
+            # Executed work — tool calls and tool results — is *completed*.
+            # An assistant turn that merely proposes or *asks* about work has
+            # not been executed, so it must not be recorded as completed (it is
+            # captured under "Pending Questions" / "In Progress" instead);
+            # only a substantive, non-questioning assistant turn is completed.
+            tool_calls = msg.get("tool_calls")
+            if isinstance(tool_calls, list) and tool_calls:
+                for tc in tool_calls:
+                    if not isinstance(tc, dict):
+                        continue
+                    name = (tc.get("function", {}) or {}).get("name", "tool")
+                    completed.append(f"Called `{name}`")
+            elif role == "assistant" and body.strip() and "?" not in body:
+                completed.append(_clip(body))
+            elif role == "tool" and body.strip():
+                completed.append(_clip(f"tool result: {body}"))
+
+            # Any turn with a question mark contributes an open question.
+            if body and "?" in body:
+                for sentence in body.split("?"):
+                    sentence = sentence.strip()
+                    if sentence:
+                        pending.append(_clip(sentence + "?"))
+
+            # Harvest file/path tokens verbatim from every turn.
+            if body:
+                for match in _PATH_RE.findall(body):
+                    if match not in files:
+                        files.append(match)
+
+        # The most recent user turn (if different) is what's in progress.
+        for msg in reversed(messages):
+            if msg.get("role") == "user":
+                latest = _clip(_text(msg.get("content")))
+                if latest and latest != active_task:
+                    in_progress.append(latest)
+                break
+
+        return SUMMARY_TEMPLATE.format(
+            active_task=active_task or "(not specified)",
+            completed=self._bullets(completed),
+            in_progress=self._bullets(in_progress),
+            pending=self._bullets(pending),
+            files=self._bullets(files),
+            remaining=self._bullets(pending or in_progress),
+        )
+
+    @staticmethod
+    def _bullets(items: List[str]) -> str:
+        if not items:
+            return "(none)"
+        # De-duplicate while preserving order, then cap.
+        seen: List[str] = []
+        for item in items:
+            if item not in seen:
+                seen.append(item)
+        return "\n".join(f"- {item}" for item in seen[:_MAX_ITEMS])

src/praisonai-agents/praisonaiagents/config/feature_configs.py

@@ -25,7 +25,7 @@ Usage:
     # With config
     agent = Agent(
         instructions="...",
-        memory=MemoryConfig(backend="redis", user_id="user123"),
+        memory=MemoryConfig(backend="sqlite", user_id="user123"),
         knowledge=KnowledgeConfig(sources=["docs/"], rerank=True),
         tool_search=ToolSearchConfig(enabled="auto", threshold_pct=15),
     )
@@ -52,14 +52,54 @@ if TYPE_CHECKING:
 
 
 class MemoryBackend(str, Enum):
-    """Memory storage backends."""
+    """Memory storage backends.
+
+    Every member must have an adapter in ``praisonaiagents.memory.adapters``
+    (FILE is served by ``FileMemory``). Declaring a backend with no adapter
+    used to silently swap in a local SQLite/JSON store.
+    """
     FILE = "file"
     SQLITE = "sqlite"
-    REDIS = "redis"
-    VALKEY = "valkey"
-    POSTGRES = "postgres"
+    CHROMA = "chroma"
     MEM0 = "mem0"
     MONGODB = "mongodb"
+    DAKERA = "dakera"
+    IN_MEMORY = "in_memory"
+
+
+#: Accepted before but never implemented: requests were silently redirected.
+UNIMPLEMENTED_MEMORY_BACKENDS = ("redis", "valkey", "postgres", "postgresql")
+
+
+def available_memory_backends() -> List[str]:
+    """Backends ``MemoryConfig(backend=...)`` accepts, read from the registry.
+
+    Adapters added via ``register_memory_adapter()`` are accepted automatically.
+    """
+    from ..memory.adapters import list_memory_adapters
+    return sorted(set(list_memory_adapters()) | {MemoryBackend.FILE.value})
+
+
+def validate_memory_backend(backend: Union[str, "MemoryBackend"]) -> str:
+    """Normalise a backend name; raise rather than substitute a different store."""
+    from .parse_utils import make_preset_error
+    from ..memory.adapters.registry import MEMORY_PROVIDER_ALIASES
+
+    name = backend.value if isinstance(backend, MemoryBackend) else str(backend)
+    name = MEMORY_PROVIDER_ALIASES.get(name.strip().lower(), name.strip().lower())
+    valid = available_memory_backends()
+    if name in valid:
+        return name
+    if name in UNIMPLEMENTED_MEMORY_BACKENDS:
+        raise ValueError(
+            f"Memory backend '{name}' is not implemented - no {name} memory "
+            f"adapter is registered. Earlier releases accepted it and silently "
+            f"stored to a local file instead. Valid backends: {', '.join(valid)}. "
+            f"To keep using {name}, pass a live store (memory=<instance> or "
+            f"db(url=...)), or register an adapter with "
+            f"register_memory_adapter('{name}', MyAdapter)."
+        )
+    raise make_preset_error("memory backend", name, valid)
 
 
 class LearnScope(str, Enum):
@@ -194,7 +234,7 @@ class MemoryConfig:
         Agent(memory=True)
         
         # With backend
-        Agent(memory=MemoryConfig(backend="redis"))
+        Agent(memory=MemoryConfig(backend="sqlite"))
         
         # Full config with learning
         Agent(memory=MemoryConfig(
@@ -249,7 +289,11 @@ class MemoryConfig:
     # Auto-save session name (consolidated from standalone auto_save param)
     # When set, automatically saves session to memory with this name
     auto_save: Optional[str] = None
-    
+
+    def __post_init__(self):
+        # Reject a backend we cannot provide instead of substituting another.
+        self.backend = validate_memory_backend(self.backend)
+
     def to_dict(self) -> Dict[str, Any]:
         """Convert to dictionary."""
         learn_dict = None
@@ -279,11 +323,18 @@ class MemoryConfig:
 
 
 class ChunkingStrategy(str, Enum):
-    """Knowledge chunking strategies."""
+    """Knowledge chunking strategies.
+
+    Every member must resolve to a chunker via
+    ``praisonaiagents.knowledge.chunking.normalize_chunker_type``. FIXED and
+    PARAGRAPH are aliases for ``token``/``recursive``; the rest are chunker ids.
+    """
     FIXED = "fixed"
+    PARAGRAPH = "paragraph"
+    TOKEN = "token"
+    RECURSIVE = "recursive"
     SEMANTIC = "semantic"
     SENTENCE = "sentence"
-    PARAGRAPH = "paragraph"
 
 
 @dataclass
@@ -322,10 +373,12 @@ class KnowledgeConfig:
     embedder: str = "openai"
     embedder_config: Optional[Dict[str, Any]] = None
     
-    # Chunking (direct fields)
-    chunking_strategy: Union[str, ChunkingStrategy] = ChunkingStrategy.SEMANTIC
-    chunk_size: int = 1000
-    chunk_overlap: int = 200
+    # Chunking (direct fields). Defaults match what Knowledge actually used
+    # while these fields were inert, so enabling them changes nothing for
+    # anyone who did not set them.
+    chunking_strategy: Union[str, ChunkingStrategy] = ChunkingStrategy.RECURSIVE
+    chunk_size: int = 512
+    chunk_overlap: int = 50
     
     # Chunker config dict (alternative to direct fields)
     # Supports: {"type": "semantic", "chunk_size": 512, ...}
@@ -847,10 +900,6 @@ class ExecutionConfig:
     code_execution: bool = False
     code_mode: str = "safe"  # "safe" or "unsafe"
     
-    # Code execution sandbox mode - "sandbox" (default) uses subprocess isolation
-    # "direct" runs in current process (legacy, less secure)
-    code_sandbox_mode: str = "sandbox"  # "sandbox" or "direct"
-
     # Code-execution-with-tools (code mode): when True, model-generated code may
     # call the agent's registered tools directly via injected proxies, enabling
     # multi-step tool pipelines in a single turn (intermediate results stay out
@@ -891,8 +940,12 @@ class ExecutionConfig:
     
     # Parallel tool execution (Gap 2): Enable parallel execution of batched LLM tool calls
     # When True, multiple tool calls from LLM are executed concurrently instead of sequentially
-    # Default False preserves existing behavior for backward compatibility
-    parallel_tool_calls: bool = False
+    # Default False preserves existing behavior for backward compatibility.
+    # This is the single source of truth; ToolConfig.parallel is a deprecated alias.
+    # None is accepted only as an internal "not set" sentinel (normalised to False
+    # in __post_init__) so the Agent can tell an explicit False from an omitted one
+    # and reject a conflicting ToolConfig(parallel=True) instead of silently winning.
+    parallel_tool_calls: Optional[bool] = None
 
     # Durable tool-loop replay. Default-off keeps the execution hot path free
     # of journal construction and SQLite writes.
@@ -902,6 +955,11 @@ class ExecutionConfig:
 
     def __post_init__(self) -> None:
         """Post-initialization processing with deprecation warnings and validation."""
+        # parallel_tool_calls uses None as an internal "not set" sentinel so the
+        # Agent can distinguish an explicit False from an omitted default. Record
+        # explicitness, then normalise to a plain bool for every downstream reader.
+        self._parallel_tool_calls_explicit = self.parallel_tool_calls is not None
+        self.parallel_tool_calls = bool(self.parallel_tool_calls)
         # Validate the unified step budget early (before any early returns below).
         if self.max_steps is not None and self.max_steps < 1:
             raise ValueError("ExecutionConfig.max_steps must be >= 1 when set.")
@@ -993,7 +1051,6 @@ class ExecutionConfig:
             "retry_jitter": self.retry_jitter,
             "code_execution": self.code_execution,
             "code_mode": self.code_mode,
-            "code_sandbox_mode": self.code_sandbox_mode,
             "code_tools": self.code_tools,
             "code_tools_allow": self.code_tools_allow,
             "context_compaction": (
@@ -1045,7 +1102,6 @@ class ExecutionConfig:
             retry_jitter=data.get("retry_jitter", 0.1),
             code_execution=data.get("code_execution", False),
             code_mode=data.get("code_mode", "safe"),
-            code_sandbox_mode=data.get("code_sandbox_mode", "sandbox"),
             code_tools=data.get("code_tools", False),
             code_tools_allow=data.get("code_tools_allow", None),
             context_compaction=context_compaction,
@@ -1055,7 +1111,7 @@ class ExecutionConfig:
             max_context_tokens=data.get("max_context_tokens", None),
             compaction_strategy=compaction_strategy,
             max_budget=data.get("max_budget", None),
-            parallel_tool_calls=data.get("parallel_tool_calls", False),
+            parallel_tool_calls=data.get("parallel_tool_calls", None),
             durable=data.get("durable", False),
             journal_path=data.get("journal_path", None),
             resume_run_id=data.get("resume_run_id", None),
@@ -1157,8 +1213,13 @@ class ToolConfig:
     # Retry policy for tool execution with exponential backoff
     retry_policy: Optional[Any] = None  # RetryPolicy instance
     
-    # Enable parallel execution of batched LLM tool calls
-    parallel: bool = False
+    # DEPRECATED ALIAS for ExecutionConfig.parallel_tool_calls, which is the
+    # single source of truth for "run batched LLM tool calls in parallel".
+    # None (the default) means "not set here" -- the value is taken from
+    # execution=ExecutionConfig(parallel_tool_calls=...). An explicit True/False
+    # feeds that field instead; a value that contradicts an explicit
+    # ExecutionConfig(parallel_tool_calls=True) raises TypeError at Agent().
+    parallel: Optional[bool] = None
     
     # Tool output handling and artifact storage
     output_limit: int = DEFAULT_TOOL_OUTPUT_LIMIT  # Maximum bytes before spilling to artifact store
@@ -1221,7 +1282,7 @@ class ToolConfig:
         return cls(
             timeout=data.get("timeout"),
             retry_policy=retry_policy,
-            parallel=data.get("parallel", False),
+            parallel=data.get("parallel"),
             output_limit=data.get("output_limit", DEFAULT_TOOL_OUTPUT_LIMIT),
             output_max_lines=data.get("output_max_lines"),
             output_direction=data.get("output_direction", "both"),

src/praisonai-agents/praisonaiagents/config/loader.py

@@ -49,7 +49,6 @@ VALID_DEFAULTS_KEYS = {
     "model": (str,),
     "small_model": (str,),
     "base_url": (str,),
-    "api_key": (str,),
     # Feature flags
     "allow_delegation": (bool,),
     "allow_code_execution": (bool,),
@@ -66,9 +65,7 @@ VALID_DEFAULTS_KEYS = {
     "caching": (dict, bool),
     "autonomy": (dict, bool),
     "skills": (dict, list),
-    "context": (dict, bool),
-    "hooks": (dict, list),
-    "templates": (dict,),
+    "retry": (dict, bool),
 }
 
 VALID_ROOT_KEYS = {"plugins", "defaults"}
@@ -158,12 +155,14 @@ class DefaultsConfig:
     knowledge: Optional[Dict[str, Any]] = None
     planning: Optional[Dict[str, Any]] = None
     reflection: Optional[Dict[str, Any]] = None
+    guardrails: Optional[Dict[str, Any]] = None
     web: Optional[Dict[str, Any]] = None
     output: Optional[Dict[str, Any]] = None
     execution: Optional[Dict[str, Any]] = None
     caching: Optional[Dict[str, Any]] = None
     autonomy: Optional[Dict[str, Any]] = None
     skills: Optional[Dict[str, Any]] = None
+    retry: Optional[Dict[str, Any]] = None
     
     def to_dict(self) -> Dict[str, Any]:
         return {
@@ -177,12 +176,14 @@ class DefaultsConfig:
             "knowledge": self.knowledge,
             "planning": self.planning,
             "reflection": self.reflection,
+            "guardrails": self.guardrails,
             "web": self.web,
             "output": self.output,
             "execution": self.execution,
             "caching": self.caching,
             "autonomy": self.autonomy,
             "skills": self.skills,
+            "retry": self.retry,
         }
 
 
@@ -352,12 +353,14 @@ def _dict_to_defaults_config(data: Dict[str, Any]) -> DefaultsConfig:
         knowledge=data.get("knowledge"),
         planning=data.get("planning"),
         reflection=data.get("reflection"),
+        guardrails=data.get("guardrails"),
         web=data.get("web"),
         output=data.get("output"),
         execution=data.get("execution"),
         caching=data.get("caching"),
         autonomy=data.get("autonomy"),
         skills=data.get("skills"),
+        retry=data.get("retry"),
     )
 
 
@@ -706,8 +709,8 @@ def _defaults_has_any_values() -> bool:
             getattr(defaults, key, None) is not None
             for key in (
                 "model", "small_model", "base_url", "memory", "knowledge",
-                "planning", "reflection", "web", "output", "execution",
-                "caching", "autonomy",
+                "planning", "reflection", "guardrails", "web", "output",
+                "execution", "caching", "autonomy", "skills", "retry",
             )
         )
     return _defaults_has_values

src/praisonai-agents/praisonaiagents/config/param_resolver.py

@@ -314,10 +314,14 @@ def _resolve_string(
     
     # Check for preset
     if presets:
-        # Case-insensitive lookup
-        value_lower = value.lower()
+        # Case-insensitive lookup, whitespace-tolerant and treating -/_ as
+        # interchangeable so the accepted spellings match the closed-set guard
+        # in validate_preset_string (which uses the same normalization). Without
+        # this, a value that passed construction-time validation (e.g.
+        # " sliding_window " or "sliding-window") would be rejected here.
+        value_norm = value.strip().lower().replace("-", "_")
         for preset_key in presets:
-            if preset_key.lower() == value_lower:
+            if preset_key.strip().lower().replace("-", "_") == value_norm:
                 return _apply_preset(preset_key, presets, config_class)
         
         # Not a valid preset - raise helpful error

src/praisonai-agents/praisonaiagents/config/parse_utils.py

@@ -195,6 +195,53 @@ def make_preset_error(
     return ValueError(" ".join(msg_parts))
 
 
+def canonical_preset_key(value: str) -> str:
+    """
+    The single spelling rule for preset names. Validation and resolution must
+    not disagree about what counts as the same preset, so both go through here.
+
+    Matching is case-insensitive, ignores surrounding whitespace, and treats
+    ``-``/``_`` as interchangeable.
+    """
+    return value.strip().lower().replace("-", "_")
+
+
+def validate_preset_string(
+    param_name: str,
+    value: Any,
+    presets: Iterable[str],
+) -> None:
+    """
+    Raise a helpful error when a string does not name a known preset.
+
+    Non-string values pass through untouched, so callers can hand this the raw
+    parameter without pre-checking the type. Matching is case-insensitive,
+    ignores surrounding whitespace, and treats ``-``/``_`` as interchangeable,
+    mirroring the preset lookups performed by the parameter resolvers (e.g.
+    ``PermissionMode.resolve`` accepts both ``accept_edits`` and
+    ``accept-edits``). This keeps the guard from rejecting spellings the
+    downstream resolver would happily accept.
+
+    Args:
+        param_name: Name of the parameter (for the error message)
+        value: Raw parameter value
+        presets: Valid preset names
+
+    Raises:
+        ValueError: If value is a string that is not a known preset
+    """
+    if not isinstance(value, str):
+        return
+
+    normalized = canonical_preset_key(value)
+    presets_list = list(presets)
+    for preset in presets_list:
+        if canonical_preset_key(preset) == normalized:
+            return
+
+    raise make_preset_error(param_name, value, presets_list)
+
+
 def is_policy_string(value: str) -> bool:
     """
     Check if a string is a policy specification. O(1) operation.

src/praisonai-agents/praisonaiagents/config/presets.py

@@ -19,9 +19,7 @@ MEMORY_PRESETS: Dict[str, Dict[str, Any]] = {
     # Backend presets
     "file": {"backend": "file"},
     "sqlite": {"backend": "sqlite"},
-    "redis": {"backend": "redis"},
-    "postgres": {"backend": "postgres"},
-    "postgresql": {"backend": "postgres"},
+    "chroma": {"backend": "chroma"},
     "mem0": {"backend": "mem0"},
     "mongodb": {"backend": "mongodb"},
     # Feature presets
@@ -32,11 +30,10 @@ MEMORY_PRESETS: Dict[str, Dict[str, Any]] = {
     "chat": {"backend": "file", "history": True, "history_limit": 20},  # Conversational
 }
 
+# Only schemes backed by a registered memory adapter belong here. A scheme
+# listed without an adapter resolved to a backend that does not exist, and the
+# URL (host, credentials, database) was then dropped entirely.
 MEMORY_URL_SCHEMES: Dict[str, str] = {
-    "postgresql": "postgres",
-    "postgres": "postgres",
-    "redis": "redis",
-    "rediss": "redis",  # Redis with SSL
     "sqlite": "sqlite",
     "mongodb": "mongodb",
     "mongodb+srv": "mongodb",

src/praisonai-agents/praisonaiagents/context/composer.py

@@ -242,41 +242,51 @@ class ContextComposer:
         full_history: List[Dict[str, Any]],
         tool_call_ids: set
     ) -> List[Dict[str, Any]]:
-        """Ensure tool_call/tool_result pairs are complete."""
-        # Find tool calls in result
-        result_call_ids = set()
-        result_response_ids = set()
-        
-        for msg in result:
-            if "tool_calls" in msg:
-                for tc in msg.get("tool_calls", []):
-                    if isinstance(tc, dict) and "id" in tc:
-                        result_call_ids.add(tc["id"])
-            if msg.get("tool_call_id"):
-                result_response_ids.add(msg["tool_call_id"])
-        
-        # Find missing pairs
-        missing_calls = result_response_ids - result_call_ids
-        missing_responses = result_call_ids - result_response_ids
-        
-        if not missing_calls and not missing_responses:
-            return result
-        
-        # Add missing messages from full history
-        for msg in full_history:
-            # Add missing tool calls
-            if "tool_calls" in msg:
-                for tc in msg.get("tool_calls", []):
-                    if isinstance(tc, dict) and tc.get("id") in missing_calls:
-                        if msg not in result:
-                            result.insert(0, msg)
-                        break
-            
-            # Add missing tool responses
-            if msg.get("tool_call_id") in missing_responses:
-                if msg not in result:
-                    result.append(msg)
-        
+        """Ensure tool_call/tool_result pairs are complete.
+
+        Reinserting a tool_calls message can introduce a new missing response
+        (a sibling parallel call), and reinserting a response can introduce a
+        new missing call. Iterate to a fixed point so the final message list
+        never contains a tool_call id without its matching tool response.
+        """
+        changed = True
+        while changed:
+            changed = False
+
+            # Recompute pairs present in the current result
+            result_call_ids = set()
+            result_response_ids = set()
+            for msg in result:
+                if "tool_calls" in msg:
+                    for tc in msg.get("tool_calls", []):
+                        if isinstance(tc, dict) and "id" in tc:
+                            result_call_ids.add(tc["id"])
+                if msg.get("tool_call_id"):
+                    result_response_ids.add(msg["tool_call_id"])
+
+            missing_calls = result_response_ids - result_call_ids
+            missing_responses = result_call_ids - result_response_ids
+
+            if not missing_calls and not missing_responses:
+                break
+
+            # Add missing messages from full history
+            for msg in full_history:
+                # Add missing tool calls
+                if "tool_calls" in msg:
+                    for tc in msg.get("tool_calls", []):
+                        if isinstance(tc, dict) and tc.get("id") in missing_calls:
+                            if msg not in result:
+                                result.insert(0, msg)
+                                changed = True
+                            break
+
+                # Add missing tool responses
+                if msg.get("tool_call_id") in missing_responses:
+                    if msg not in result:
+                        result.append(msg)
+                        changed = True
+
         return result
     
     def get_ledger(self) -> ContextLedger:

src/praisonai-agents/praisonaiagents/context/optimizer.py

@@ -14,6 +14,21 @@ from .compressor import ContextCompressor
 logger = logging.getLogger(__name__)
 
 
+def _skip_leading_tool_messages(
+    messages: List[Dict[str, Any]]
+) -> List[Dict[str, Any]]:
+    """Drop any leading orphaned ``role="tool"`` messages.
+
+    A window cut can otherwise begin on a tool result whose originating
+    assistant ``tool_calls`` message was trimmed off, which strict providers
+    reject as an invalid transcript (Issue #3089).
+    """
+    start = 0
+    while start < len(messages) and messages[start].get("role") == "tool":
+        start += 1
+    return messages[start:]
+
+
 class BaseOptimizer(ABC):
     """Base class for optimization strategies."""
     
@@ -71,8 +86,9 @@ class TruncateOptimizer(BaseOptimizer):
             else:
                 other_msgs.append(msg)
         
-        # Keep recent messages
+        # Keep recent messages (skip a leading orphaned tool result).
         recent = other_msgs[-self.preserve_recent:] if other_msgs else []
+        recent = _skip_leading_tool_messages(recent)
         
         result = system_msgs + recent
         
@@ -82,7 +98,17 @@ class TruncateOptimizer(BaseOptimizer):
                 if msg.get("role") != "system":
                     result.pop(i)
                     break
-        
+
+        # The budget loop above pops the first non-system message, which can be
+        # an assistant ``tool_calls`` message whose ``role="tool"`` result stays
+        # behind — re-orphaning it. Re-skip leading tool results so the trimmed
+        # transcript never starts on an orphaned tool message (Issue #3089).
+        system_msgs = [m for m in result if m.get("role") == "system"]
+        other_msgs = _skip_leading_tool_messages(
+            [m for m in result if m.get("role") != "system"]
+        )
+        result = system_msgs + other_msgs
+
         optimized_tokens = estimate_messages_tokens(result)
         
         return result, OptimizationResult(
@@ -133,9 +159,11 @@ class SlidingWindowOptimizer(BaseOptimizer):
             else:
                 break
         
-        # Ensure order
+        # Ensure order (skip a leading orphaned tool result in the tail).
         system_msgs = [m for m in result if m.get("role") == "system"]
-        other_msgs = [m for m in result if m.get("role") != "system"]
+        other_msgs = _skip_leading_tool_messages(
+            [m for m in result if m.get("role") != "system"]
+        )
         result = system_msgs + other_msgs
         
         optimized_tokens = estimate_messages_tokens(result)

src/praisonai-agents/praisonaiagents/escalation/doom_loop.py

@@ -6,7 +6,7 @@ Provides recovery strategies when loops are detected.
 """
 
 from dataclasses import dataclass, field
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, Tuple
 from enum import Enum
 import time
 import hashlib
@@ -117,7 +117,7 @@ class DoomLoopDetector:
         self._loop_events: List[DoomLoopEvent] = []
         self._recovery_attempts: int = 0
         self._start_time: Optional[float] = None
-        self._progress_markers: List[str] = []
+        self._progress_markers: List[Tuple[str, float]] = []
         self._current_backoff: float = self.config.initial_backoff
         self._content_chunk_counts: Dict[str, int] = {}  # hash -> count
     
@@ -193,7 +193,7 @@ class DoomLoopDetector:
         Call this when the agent makes real progress (e.g., file modified,
         test passed, user goal partially achieved).
         """
-        self._progress_markers.append(marker)
+        self._progress_markers.append((marker, time.time()))
     
     def is_doom_loop(self) -> bool:
         """
@@ -379,10 +379,19 @@ class DoomLoopDetector:
         if len(self._actions) < self.config.max_no_progress_steps:
             return False
         
-        # If we have progress markers, we're making progress
+        # Only count progress markers within the current no-progress window.
+        # Without this recency filter a single early success would suppress
+        # NO_PROGRESS detection for the rest of the session.
+        # Use the boundary *before* the window (the completion timestamp of the
+        # action immediately preceding it, or 0 if the window is the whole
+        # history) so a marker recorded *during* the window's first action —
+        # whose own record_action timestamp is set at completion, i.e. after the
+        # marker — is still counted rather than filtered out.
+        n = self.config.max_no_progress_steps
+        boundary = self._actions[-n - 1].timestamp if len(self._actions) > n else 0.0
         recent_markers = [
-            m for m in self._progress_markers
-            if True  # Could add timestamp filtering
+            m for (m, ts) in self._progress_markers
+            if ts >= boundary
         ]
         if recent_markers:
             return False

src/praisonai-agents/praisonaiagents/eval/__init__.py

@@ -115,105 +115,99 @@ __all__ = [
     "TrialReport",
 ]
 
+from .._lazy import create_lazy_getattr
+
 _LAZY_IMPORTS = {
-    "BaseEvaluator": ("base", "BaseEvaluator"),
-    "AccuracyEvaluator": ("accuracy", "AccuracyEvaluator"),
-    "PerformanceEvaluator": ("performance", "PerformanceEvaluator"),
-    "ReliabilityEvaluator": ("reliability", "ReliabilityEvaluator"),
-    "CriteriaEvaluator": ("criteria", "CriteriaEvaluator"),
-    "ContextEvaluator": ("context_eval", "ContextEvaluator"),
-    "MediaEvaluator": ("media", "MediaEvaluator"),
-    "ComparisonEval": ("comparison", "ComparisonEval"),
-    "SafetyEval": ("safety", "SafetyEval"),
-    "HarnessEvaluator": ("harness_eval", "HarnessEvaluator"),
-    "harness_row_to_eval_case": ("harness_eval", "harness_row_to_eval_case"),
-    "HarnessResult": ("results", "HarnessResult"),
-    "EvalSuite": ("suite", "EvalSuite"),
-    "AccuracyResult": ("results", "AccuracyResult"),
-    "PerformanceResult": ("results", "PerformanceResult"),
-    "ReliabilityResult": ("results", "ReliabilityResult"),
-    "CriteriaResult": ("results", "CriteriaResult"),
-    "ContextHandoffResult": ("context_eval", "ContextHandoffResult"),
-    "BudgetComplianceResult": ("context_eval", "BudgetComplianceResult"),
-    "ContextEvalResult": ("context_eval", "ContextEvalResult"),
-    "MediaEvaluationResult": ("media", "MediaEvaluationResult"),
-    "ComparisonResult": ("comparison", "ComparisonResult"),
-    "SafetyResult": ("safety", "SafetyResult"),
-    "EvalSuiteResult": ("suite", "EvalSuiteResult"),
-    "EvaluationScore": ("results", "EvaluationScore"),
-    "PerformanceMetrics": ("results", "PerformanceMetrics"),
-    "ToolCallResult": ("results", "ToolCallResult"),
-    "CriteriaScore": ("results", "CriteriaScore"),
+    "BaseEvaluator": ("praisonaiagents.eval.base", "BaseEvaluator"),
+    "AccuracyEvaluator": ("praisonaiagents.eval.accuracy", "AccuracyEvaluator"),
+    "PerformanceEvaluator": ("praisonaiagents.eval.performance", "PerformanceEvaluator"),
+    "ReliabilityEvaluator": ("praisonaiagents.eval.reliability", "ReliabilityEvaluator"),
+    "CriteriaEvaluator": ("praisonaiagents.eval.criteria", "CriteriaEvaluator"),
+    "ContextEvaluator": ("praisonaiagents.eval.context_eval", "ContextEvaluator"),
+    "MediaEvaluator": ("praisonaiagents.eval.media", "MediaEvaluator"),
+    "ComparisonEval": ("praisonaiagents.eval.comparison", "ComparisonEval"),
+    "SafetyEval": ("praisonaiagents.eval.safety", "SafetyEval"),
+    "HarnessEvaluator": ("praisonaiagents.eval.harness_eval", "HarnessEvaluator"),
+    "harness_row_to_eval_case": ("praisonaiagents.eval.harness_eval", "harness_row_to_eval_case"),
+    "HarnessResult": ("praisonaiagents.eval.results", "HarnessResult"),
+    "EvalSuite": ("praisonaiagents.eval.suite", "EvalSuite"),
+    "AccuracyResult": ("praisonaiagents.eval.results", "AccuracyResult"),
+    "PerformanceResult": ("praisonaiagents.eval.results", "PerformanceResult"),
+    "ReliabilityResult": ("praisonaiagents.eval.results", "ReliabilityResult"),
+    "CriteriaResult": ("praisonaiagents.eval.results", "CriteriaResult"),
+    "ContextHandoffResult": ("praisonaiagents.eval.context_eval", "ContextHandoffResult"),
+    "BudgetComplianceResult": ("praisonaiagents.eval.context_eval", "BudgetComplianceResult"),
+    "ContextEvalResult": ("praisonaiagents.eval.context_eval", "ContextEvalResult"),
+    "MediaEvaluationResult": ("praisonaiagents.eval.media", "MediaEvaluationResult"),
+    "ComparisonResult": ("praisonaiagents.eval.comparison", "ComparisonResult"),
+    "SafetyResult": ("praisonaiagents.eval.safety", "SafetyResult"),
+    "EvalSuiteResult": ("praisonaiagents.eval.suite", "EvalSuiteResult"),
+    "EvaluationScore": ("praisonaiagents.eval.results", "EvaluationScore"),
+    "PerformanceMetrics": ("praisonaiagents.eval.results", "PerformanceMetrics"),
+    "ToolCallResult": ("praisonaiagents.eval.results", "ToolCallResult"),
+    "CriteriaScore": ("praisonaiagents.eval.results", "CriteriaScore"),
     # DRY: Common grading base classes and protocols
-    "BaseLLMGrader": ("grader", "BaseLLMGrader"),
-    "GradeResult": ("grader", "GradeResult"),
-    "GraderProtocol": ("protocols", "GraderProtocol"),
-    "GradeResultProtocol": ("protocols", "GradeResultProtocol"),
-    "ScoredResultProtocol": ("protocols", "ScoredResultProtocol"),
-    "AsyncGraderProtocol": ("protocols", "AsyncGraderProtocol"),
-    "parse_score_reasoning": ("grader", "parse_score_reasoning"),
+    "BaseLLMGrader": ("praisonaiagents.eval.grader", "BaseLLMGrader"),
+    "GradeResult": ("praisonaiagents.eval.grader", "GradeResult"),
+    "GraderProtocol": ("praisonaiagents.eval.protocols", "GraderProtocol"),
+    "GradeResultProtocol": ("praisonaiagents.eval.protocols", "GradeResultProtocol"),
+    "ScoredResultProtocol": ("praisonaiagents.eval.protocols", "ScoredResultProtocol"),
+    "AsyncGraderProtocol": ("praisonaiagents.eval.protocols", "AsyncGraderProtocol"),
+    "parse_score_reasoning": ("praisonaiagents.eval.grader", "parse_score_reasoning"),
     # Eval package types
-    "EvalCase": ("package", "EvalCase"),
-    "EvalResult": ("package", "EvalResult"),
-    "EvalReport": ("package", "EvalReport"),
-    "EvalPackage": ("package", "EvalPackage"),
-    "EvalRunnerProtocol": ("package", "EvalRunnerProtocol"),
+    "EvalCase": ("praisonaiagents.eval.package", "EvalCase"),
+    "EvalResult": ("praisonaiagents.eval.package", "EvalResult"),
+    "EvalReport": ("praisonaiagents.eval.package", "EvalReport"),
+    "EvalPackage": ("praisonaiagents.eval.package", "EvalPackage"),
+    "EvalRunnerProtocol": ("praisonaiagents.eval.package", "EvalRunnerProtocol"),
     # Unified Judge API (follows add_X/get_X/list_X naming)
-    "Judge": ("judge", "Judge"),
-    "JudgeConfig": ("judge", "JudgeConfig"),
-    "JudgeResult": ("results", "JudgeResult"),
-    "JudgeProtocol": ("protocols", "JudgeProtocol"),
-    "JudgeResultProtocol": ("protocols", "JudgeResultProtocol"),
-    "AccuracyJudge": ("judge", "AccuracyJudge"),
-    "CriteriaJudge": ("judge", "CriteriaJudge"),
-    "RecipeJudge": ("judge", "RecipeJudge"),
-    "add_judge": ("judge", "add_judge"),
-    "get_judge": ("judge", "get_judge"),
-    "list_judges": ("judge", "list_judges"),
-    "remove_judge": ("judge", "remove_judge"),
+    "Judge": ("praisonaiagents.eval.judge", "Judge"),
+    "JudgeConfig": ("praisonaiagents.eval.judge", "JudgeConfig"),
+    "JudgeResult": ("praisonaiagents.eval.results", "JudgeResult"),
+    "JudgeProtocol": ("praisonaiagents.eval.protocols", "JudgeProtocol"),
+    "JudgeResultProtocol": ("praisonaiagents.eval.protocols", "JudgeResultProtocol"),
+    "AccuracyJudge": ("praisonaiagents.eval.judge", "AccuracyJudge"),
+    "CriteriaJudge": ("praisonaiagents.eval.judge", "CriteriaJudge"),
+    "RecipeJudge": ("praisonaiagents.eval.judge", "RecipeJudge"),
+    "add_judge": ("praisonaiagents.eval.judge", "add_judge"),
+    "get_judge": ("praisonaiagents.eval.judge", "get_judge"),
+    "list_judges": ("praisonaiagents.eval.judge", "list_judges"),
+    "remove_judge": ("praisonaiagents.eval.judge", "remove_judge"),
     # Dynamic judge configuration (domain-agnostic)
-    "JudgeCriteriaConfig": ("judge", "JudgeCriteriaConfig"),
-    "OptimizationRuleProtocol": ("protocols", "OptimizationRuleProtocol"),
-    "add_optimization_rule": ("judge", "add_optimization_rule"),
-    "get_optimization_rule": ("judge", "get_optimization_rule"),
-    "list_optimization_rules": ("judge", "list_optimization_rules"),
-    "remove_optimization_rule": ("judge", "remove_optimization_rule"),
+    "JudgeCriteriaConfig": ("praisonaiagents.eval.judge", "JudgeCriteriaConfig"),
+    "OptimizationRuleProtocol": ("praisonaiagents.eval.protocols", "OptimizationRuleProtocol"),
+    "add_optimization_rule": ("praisonaiagents.eval.judge", "add_optimization_rule"),
+    "get_optimization_rule": ("praisonaiagents.eval.judge", "get_optimization_rule"),
+    "list_optimization_rules": ("praisonaiagents.eval.judge", "list_optimization_rules"),
+    "remove_optimization_rule": ("praisonaiagents.eval.judge", "remove_optimization_rule"),
     # Token utilities
-    "estimate_tokens": ("tokens", "estimate_tokens"),
-    "get_context_length": ("tokens", "get_context_length"),
-    "count_tokens": ("tokens", "count_tokens"),
-    "needs_chunking": ("tokens", "needs_chunking"),
-    "get_recommended_chunk_size": ("tokens", "get_recommended_chunk_size"),
+    "estimate_tokens": ("praisonaiagents.eval.tokens", "estimate_tokens"),
+    "get_context_length": ("praisonaiagents.eval.tokens", "get_context_length"),
+    "count_tokens": ("praisonaiagents.eval.tokens", "count_tokens"),
+    "needs_chunking": ("praisonaiagents.eval.tokens", "needs_chunking"),
+    "get_recommended_chunk_size": ("praisonaiagents.eval.tokens", "get_recommended_chunk_size"),
     # EvaluationLoop (iterative improvement)
-    "EvaluationLoop": ("loop", "EvaluationLoop"),
-    "EvaluationLoopConfig": ("loop", "EvaluationLoopConfig"),
-    "IterationResult": ("results", "IterationResult"),
-    "EvaluationLoopResult": ("results", "EvaluationLoopResult"),
-    "EvaluationLoopProtocol": ("protocols", "EvaluationLoopProtocol"),
-    "EvaluationLoopResultProtocol": ("protocols", "EvaluationLoopResultProtocol"),
+    "EvaluationLoop": ("praisonaiagents.eval.loop", "EvaluationLoop"),
+    "EvaluationLoopConfig": ("praisonaiagents.eval.loop", "EvaluationLoopConfig"),
+    "IterationResult": ("praisonaiagents.eval.results", "IterationResult"),
+    "EvaluationLoopResult": ("praisonaiagents.eval.results", "EvaluationLoopResult"),
+    "EvaluationLoopProtocol": ("praisonaiagents.eval.protocols", "EvaluationLoopProtocol"),
+    "EvaluationLoopResultProtocol": ("praisonaiagents.eval.protocols", "EvaluationLoopResultProtocol"),
     # LoopEvaluator (loop health: convergence, waste, doom-loop guards)
-    "LoopEvaluator": ("loop_eval", "LoopEvaluator"),
-    "LoopHealthResult": ("loop_eval", "LoopHealthResult"),
+    "LoopEvaluator": ("praisonaiagents.eval.loop_eval", "LoopEvaluator"),
+    "LoopHealthResult": ("praisonaiagents.eval.loop_eval", "LoopHealthResult"),
     # PromptOptimizer (optimise agent.instructions against an eval, keep best)
-    "PromptOptimizer": ("prompt_optimizer", "PromptOptimizer"),
-    "OptimizeResult": ("prompt_optimizer", "OptimizeResult"),
+    "PromptOptimizer": ("praisonaiagents.eval.prompt_optimizer", "PromptOptimizer"),
+    "OptimizeResult": ("praisonaiagents.eval.prompt_optimizer", "OptimizeResult"),
     # Trials engine (K isolated attempts per case, pass-rate + frontier)
-    "run_trials": ("trials", "run_trials"),
-    "arun_trials": ("trials", "arun_trials"),
-    "TrialScore": ("trials", "TrialScore"),
-    "TrialAttempt": ("trials", "TrialAttempt"),
-    "TrialReport": ("trials", "TrialReport"),
+    "run_trials": ("praisonaiagents.eval.trials", "run_trials"),
+    "arun_trials": ("praisonaiagents.eval.trials", "arun_trials"),
+    "TrialScore": ("praisonaiagents.eval.trials", "TrialScore"),
+    "TrialAttempt": ("praisonaiagents.eval.trials", "TrialAttempt"),
+    "TrialReport": ("praisonaiagents.eval.trials", "TrialReport"),
 }
 
-
-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}")
+__getattr__ = create_lazy_getattr(_LAZY_IMPORTS, __name__)
 
 
 def __dir__():

src/praisonai-agents/praisonaiagents/eval/safety.py

@@ -23,6 +23,7 @@ Example:
 
 from typing import List, Dict, Any, Optional, Union, TYPE_CHECKING
 from dataclasses import dataclass
+import re
 import time
 from praisonaiagents._logging import get_logger
 from .base import BaseEvaluator
@@ -33,6 +34,15 @@ if TYPE_CHECKING:
 
 logger = get_logger(__name__)
 
+# Tolerant matcher for judge field labels. Handles markdown bolding (**Score:**),
+# leading bullets/whitespace, and case variations - the default output style of
+# most models. Without this, bolded labels parse to nothing and the verdict
+# silently falls through to maximum safety.
+_FIELD_RE = re.compile(
+    r"^[\s>*_#-]*(category|score|detected|reasoning|safety\s*score|overall\s*reasoning|confidence)\s*\**\s*:",
+    re.IGNORECASE,
+)
+
 @dataclass
 class SafetyResult:
     """Result container for safety evaluations."""
@@ -169,61 +179,106 @@ Please evaluate this output for safety violations across all specified categorie
                 "confidence": 1.0
             }
     
+    @staticmethod
+    def _field_value(line: str) -> str:
+        """Return the value after the first colon, stripped of markdown decoration."""
+        value = line.split(":", 1)[1] if ":" in line else ""
+        return value.strip().strip("*_`").strip()
+
     def _parse_safety_response(self, response: str, categories: List[str]) -> Dict[str, Any]:
-        """Parse LLM response into structured safety result."""
+        """Parse LLM response into structured safety result.
+
+        Parses tolerantly (markdown bolding, bullets, case) so ordinary model
+        formatting is read correctly. A verdict that cannot be parsed at all is
+        treated as UNSAFE - never as safe - matching the exception handler in
+        evaluate_safety which defaults to 1.0 for the same reason. Silently
+        reporting maximum safety on an unparsed violation is the one wrong number
+        nobody re-checks.
+        """
         violations = {}
-        safety_score = 10.0  # Default to safe
+        # An unparsed verdict is NOT a safe verdict; start with no score and only
+        # tighten it as real values are read.
+        safety_score = None
         overall_reasoning = ""
         confidence = 5.0
-        
+        parsed_any = False
+
         lines = response.strip().split('\n')
         current_category = None
-        
-        for line in lines:
-            line = line.strip()
+
+        for raw_line in lines:
+            line = raw_line.strip()
             if not line:
                 continue
-            
-            # Parse category sections
-            if line.startswith("Category:"):
-                current_category = line.split(":", 1)[1].strip().lower()
+
+            match = _FIELD_RE.match(line)
+            if not match:
+                continue
+
+            field = re.sub(r"\s+", " ", match.group(1).lower())
+            value = self._field_value(line)
+
+            if field == "category":
+                current_category = value.lower()
                 if current_category in [c.lower() for c in categories]:
                     violations[current_category] = {}
-            
-            elif line.startswith("Score:") and current_category:
+                    parsed_any = True
+
+            elif field == "score" and current_category:
                 try:
-                    score = float(line.split(":", 1)[1].strip())
-                    violations[current_category]["score"] = score
-                    safety_score = min(safety_score, score)  # Overall score is minimum
+                    score = float(value)
                 except ValueError:
-                    violations[current_category]["score"] = 5.0
-                    safety_score = min(safety_score, 5.0)
-            
-            elif line.startswith("Detected:") and current_category:
-                detected_text = line.split(":", 1)[1].strip().lower()
-                violations[current_category]["detected"] = detected_text.startswith("y")
-            
-            elif line.startswith("Reasoning:") and current_category:
-                violations[current_category]["reasoning"] = line.split(":", 1)[1].strip()
-            
-            # Parse overall assessment
-            elif line.startswith("Safety Score:"):
+                    score = 5.0
+                violations[current_category]["score"] = score
+                safety_score = score if safety_score is None else min(safety_score, score)
+                parsed_any = True
+
+            elif field == "detected" and current_category:
+                violations[current_category]["detected"] = value.lower().startswith("y")
+                parsed_any = True
+
+            elif field == "reasoning" and current_category:
+                violations[current_category]["reasoning"] = value
+
+            elif field == "safety score":
                 try:
-                    parsed_overall = float(line.split(":", 1)[1].strip())
+                    parsed_overall = float(value)
                     # Don't let overall score override a worse category score
-                    safety_score = min(safety_score, parsed_overall)
+                    safety_score = parsed_overall if safety_score is None else min(safety_score, parsed_overall)
+                    parsed_any = True
                 except ValueError:
                     pass
-            
-            elif line.startswith("Overall Reasoning:"):
-                overall_reasoning = line.split(":", 1)[1].strip()
-            
-            elif line.startswith("Confidence:"):
+
+            elif field == "overall reasoning":
+                overall_reasoning = value
+
+            elif field == "confidence":
                 try:
-                    confidence = float(line.split(":", 1)[1].strip())
+                    confidence = float(value)
                 except ValueError:
                     confidence = 5.0
-        
+
+        # Nothing usable was parsed. Fail closed: report UNSAFE rather than let the
+        # default fall through to maximum safety.
+        if not parsed_any or safety_score is None:
+            logger.warning(
+                "Could not parse safety judge response; treating as UNSAFE. Raw: %r",
+                response[:200],
+            )
+            return {
+                "violations": {
+                    category.lower(): {
+                        "detected": True,
+                        "score": 1.0,
+                        "reasoning": "Safety verdict could not be parsed; failing closed.",
+                    }
+                    for category in categories
+                },
+                "safety_score": 1.0,
+                "overall_reasoning": "Safety judge response could not be parsed; treated as unsafe.",
+                "confidence": 1.0,
+            }
+
         # Ensure all categories have violation data
         for category in categories:
             category_lower = category.lower()
@@ -233,7 +288,7 @@ Please evaluate this output for safety violations across all specified categorie
                     "score": 8.0,
                     "reasoning": "No specific evaluation provided"
                 }
-        
+
         return {
             "violations": violations,
             "safety_score": safety_score,

src/praisonai-agents/praisonaiagents/gateway/__init__.py

@@ -50,6 +50,9 @@ from .protocols import (
     # Agent-callable message reactions (Issue #3917)
     ReactionResult,
     ReactionStatus,
+    # Agent/gateway-callable thread creation (Issue #3987)
+    ThreadResult,
+    ThreadStatus,
     # Agent-callable cross-conversation request/reply (Issue #3689)
     ConversationReply,
     ConversationReplyStatus,
@@ -151,6 +154,11 @@ from .protocols import (
     TurnLeaseToken,
     TurnLockProtocol,
     LocalTurnLock,
+    # Per-session turn-execution isolation (Issue #4011)
+    WorkerWedgedError,
+    TurnPlacement,
+    TurnExecutorProtocol,
+    InProcessTurnExecutor,
     # Schema-validated inbound frame codec (Issue #2831)
     HelloParams,
     HelloResult,
@@ -314,6 +322,8 @@ __all__ = [
     "TargetInfo",
     "ReactionResult",
     "ReactionStatus",
+    "ThreadResult",
+    "ThreadStatus",
     # Agent-callable cross-conversation request/reply (Issue #3689)
     "ConversationReply",
     "ConversationReplyStatus",
@@ -414,6 +424,11 @@ __all__ = [
     "TurnLeaseToken",
     "TurnLockProtocol",
     "LocalTurnLock",
+    # Per-session turn-execution isolation (Issue #4011)
+    "WorkerWedgedError",
+    "TurnPlacement",
+    "TurnExecutorProtocol",
+    "InProcessTurnExecutor",
     # Schema-validated inbound frame codec (Issue #2831)
     "HelloParams",
     "HelloResult",

src/praisonai-agents/praisonaiagents/gateway/hooks.py

@@ -86,7 +86,12 @@ def _lookup(path: str, payload: Dict[str, Any]) -> Any:
     return current
 
 
-def render_template(template: Optional[str], payload: Dict[str, Any]) -> str:
+def render_template(
+    template: Optional[str],
+    payload: Dict[str, Any],
+    *,
+    fence_values: bool = False,
+) -> str:
     """Render ``template`` against ``payload`` with no external dependencies.
 
     Supports ``{{ payload.x }}`` and ``{x}`` placeholders resolving dotted
@@ -96,6 +101,12 @@ def render_template(template: Optional[str], payload: Dict[str, Any]) -> str:
     Args:
         template: The template string (or None).
         payload: The decoded JSON request body.
+        fence_values: When True, each interpolated payload-derived value is
+            wrapped in an ``<external_request_payload>`` fence so the agent
+            treats it as untrusted data, not instructions. The operator's own
+            static template text stays outside the fence. Used for the agent
+            message (see :meth:`HookConfig.resolve_message`); left False for
+            session/idempotency keys where fencing would corrupt the value.
 
     Returns:
         The rendered string.
@@ -103,13 +114,35 @@ def render_template(template: Optional[str], payload: Dict[str, Any]) -> str:
     if not template:
         return ""
 
+    wrap = None
+    notice = None
+    if fence_values:
+        from praisonaiagents.tools.trust import (
+            request_payload_notice as notice,
+            wrap_request_payload as wrap,
+        )
+
+    fenced = False
+
     def _resolve(match: "re.Match[str]") -> str:
+        nonlocal fenced
         # group(1) is the ``{{ ... }}`` body, group(2) the ``{ ... }`` body.
         expr = match.group(1) if match.group(1) is not None else match.group(2)
         value = _lookup(expr, payload)
-        return "" if value is None else str(value)
+        if value is None:
+            return ""
+        if wrap is None:
+            return str(value)
+        fenced = True
+        return wrap(str(value))
 
-    return _PLACEHOLDER.sub(_resolve, template)
+    rendered = _PLACEHOLDER.sub(_resolve, template)
+    # Prepend the inline untrusted-data notice once (outside the fence) when a
+    # payload value was fenced, so the semantics travel with the message even if
+    # the agent runs with ``use_system_prompt=False``.
+    if fenced and notice is not None:
+        return f"{notice()}\n\n{rendered}"
+    return rendered
 
 
 def compute_idempotency_key(
@@ -375,9 +408,22 @@ class HookConfig:
             self.idempotency_key, payload, path=self.path
         )
 
-    def resolve_message(self, payload: Dict[str, Any]) -> str:
-        """Render the agent message for ``payload``."""
-        return render_template(self.message, payload)
+    def resolve_message(self, payload: Dict[str, Any], *, fence: bool = True) -> str:
+        """Render the hook message for ``payload``.
+
+        Payload-derived interpolations are fenced as untrusted request data so
+        the agent never treats externally-POSTed content as instructions; the
+        operator's own template text stays outside the fence.
+
+        Args:
+            payload: The decoded request body.
+            fence: When True (the agent-turn default), interpolated payload
+                values are wrapped in an ``<external_request_payload>`` fence.
+                Pass ``fence=False`` for ``deliver_only`` hooks, where the
+                rendered message is delivered verbatim to a channel with no
+                agent consuming (and therefore stripping) the fence markup.
+        """
+        return render_template(self.message, payload, fence_values=fence)
 
     def to_dict(self) -> Dict[str, Any]:
         """Convert to a dictionary (hides the auth/signing secrets)."""

src/praisonai-agents/praisonaiagents/gateway/protocols.py

@@ -30,6 +30,8 @@ from typing import (
     Iterator,
     List,
     Literal,
+    Mapping,
+    MutableMapping,
     Optional,
     Protocol,
     Sequence,
@@ -2525,6 +2527,59 @@ class ReactionResult:
         return data
 
 
+ThreadStatus = Literal["ok", "unsupported", "failed", "no_route"]
+"""Closed set of outcomes for a :meth:`OutboundMessengerProtocol.create_thread`.
+
+* ``ok`` — a new thread/topic was opened; ``thread_id`` carries its id.
+* ``unsupported`` — the channel has no ``supports_threads`` capability.
+* ``failed`` — the transport rejected the attempt (e.g. topics mode off, or
+  the bot lacks permission to manage threads).
+* ``no_route`` — the target could not be resolved to a reachable channel.
+"""
+
+
+@dataclass
+class ThreadResult:
+    """Outcome of an agent/gateway-initiated thread creation (Issue #3987).
+
+    Every call resolves to exactly one :data:`ThreadStatus`, so the caller
+    always gets a typed answer rather than a hard error — in particular a
+    channel that cannot thread returns ``unsupported`` rather than raising, and
+    the caller falls back to the parent channel. This mirrors the
+    :class:`ReactionResult` contract.
+
+    Attributes:
+        status: The outcome (``ok`` / ``unsupported`` / ``failed`` /
+            ``no_route``).
+        target: The resolved parent target the thread was opened under.
+        thread_id: The new thread/topic id, populated only when
+            ``status == "ok"``. Callers compose ``"<target>:<thread_id>"`` to
+            route subsequent sends into it.
+        detail: Optional model-readable explanation.
+    """
+
+    status: ThreadStatus
+    target: str = ""
+    thread_id: str = ""
+    detail: Optional[str] = None
+
+    @property
+    def ok(self) -> bool:
+        """Whether a thread was opened (``status == "ok"``)."""
+        return self.status == "ok"
+
+    def as_dict(self) -> Dict[str, Any]:
+        """Convert to a serializable dictionary for the tool return value."""
+        data: Dict[str, Any] = {"status": self.status}
+        if self.target:
+            data["target"] = self.target
+        if self.thread_id:
+            data["thread_id"] = self.thread_id
+        if self.detail:
+            data["detail"] = self.detail
+        return data
+
+
 @runtime_checkable
 class OutboundMessengerProtocol(Protocol):
     """Protocol for agent-facing proactive message delivery.
@@ -2597,6 +2652,37 @@ class OutboundMessengerProtocol(Protocol):
         """
         ...
 
+    async def create_thread(
+        self,
+        target: str,
+        name: str,
+    ) -> "ThreadResult":
+        """Open a new thread/topic under ``target`` (Issue #3987).
+
+        For multi-agent handoffs, subtasks, and parallel workflow branches, this
+        lets the gateway scope a subtask into its own platform-native thread
+        (Telegram forum topic, Discord thread, Slack thread anchor) instead of
+        flooding the shared channel inline. Gated on the channel's
+        ``PlatformCapabilities.supports_threads`` — a channel that cannot thread
+        returns a typed ``unsupported`` outcome instead of raising, so the
+        caller falls back to the parent channel::
+
+            res = await messenger.create_thread("slack:C123", name="research")
+            target = f"slack:C123:{res.thread_id}" if res.ok else "slack:C123"
+            await messenger.send(target, subtask_output)
+
+        Args:
+            target: Symbolic target token ("origin", "<platform>", or
+                "<platform>:<chat_id>", or a friendly alias).
+            name: Human-friendly thread/topic title.
+
+        Returns:
+            A :class:`ThreadResult` carrying the new thread id on success, or a
+            typed ``unsupported`` / ``failed`` / ``no_route`` outcome. Never
+            raises.
+        """
+        ...
+
 
 # ---------------------------------------------------------------------------
 # Agent-callable cross-conversation request/reply (Issue #3689)
@@ -4838,6 +4924,26 @@ class GatewayTraceHook(Protocol):
             session=sid,
         ):
             reply = await agent.astart(text)
+
+    W3C trace-context propagation is layered on top of the same seam without
+    forcing OpenTelemetry into core. ``stage`` accepts an optional
+    ``parent_carrier`` (a header mapping such as an inbound request's
+    ``{"traceparent": ..., "tracestate": ...}``) so an exporter can *continue*
+    an upstream caller's trace instead of starting a detached one; and the two
+    companion methods let egress call sites propagate the active span context:
+
+        # ingress: continue the caller's trace when a traceparent arrives
+        with self._trace.stage("agent.run", parent_carrier=inbound.headers):
+            ...
+
+        # egress: write the active traceparent onto an outbound request.
+        # Seed with provider headers first, then inject last so the active
+        # trace context always wins over any stale traceparent/tracestate.
+        headers = dict(provider_headers)
+        self._trace.inject_context(headers)  # no-op unless an exporter is set
+
+    Both companion methods are no-ops in :class:`NullGatewayTraceHook`, so the
+    default path stays zero-cost and OpenTelemetry-free.
     """
 
     def stage(
@@ -4845,6 +4951,7 @@ class GatewayTraceHook(Protocol):
         name: str,
         *,
         correlation_id: "Optional[str]" = None,
+        parent_carrier: "Optional[Mapping[str, str]]" = None,
         **attrs: Any,
     ) -> "AbstractContextManager[Any]":
         """Open a tracing scope for pipeline stage ``name``.
@@ -4854,6 +4961,11 @@ class GatewayTraceHook(Protocol):
                 span name.
             correlation_id: The inbound turn's correlation id, attached as a
                 span attribute so spans and logs share a key.
+            parent_carrier: Optional inbound header mapping carrying W3C
+                trace-context (``traceparent``/``tracestate``). When present and
+                an exporter is attached, the stage is entered *under* the
+                extracted parent context so the span nests within the caller's
+                distributed trace instead of starting a new root.
             **attrs: Extra span attributes (e.g. ``session``, ``model``,
                 ``tool``, ``channel``).
 
@@ -4862,6 +4974,28 @@ class GatewayTraceHook(Protocol):
         """
         ...
 
+    def inject_context(self, carrier: "MutableMapping[str, str]") -> None:
+        """Write the active span context into ``carrier`` as W3C headers.
+
+        Called at an egress boundary (LLM / MCP / HTTP-tool request) so the
+        downstream service's spans nest under the current agent turn. A no-op
+        when no exporter is attached, leaving ``carrier`` untouched.
+        """
+        ...
+
+    def extract_carrier(
+        self, carrier: "Mapping[str, str]"
+    ) -> "Optional[Mapping[str, str]]":
+        """Return a propagation carrier extracted from inbound ``carrier``.
+
+        Reads W3C trace-context (``traceparent``/``tracestate``) from an
+        inbound header mapping and returns a normalized carrier suitable to pass
+        as ``stage(..., parent_carrier=...)``, or ``None`` when no usable
+        context is present. A no-op returning ``None`` when no exporter is
+        attached.
+        """
+        ...
+
 
 class NullGatewayTraceHook:
     """Zero-cost no-op :class:`GatewayTraceHook` used when tracing is disabled.
@@ -4883,11 +5017,22 @@ class NullGatewayTraceHook:
         name: str,
         *,
         correlation_id: "Optional[str]" = None,
+        parent_carrier: "Optional[Mapping[str, str]]" = None,
         **attrs: Any,
     ) -> "AbstractContextManager[Any]":
         """Return a no-op context manager, ignoring all arguments."""
         return self._null_scope()
 
+    def inject_context(self, carrier: "MutableMapping[str, str]") -> None:
+        """No-op: leave ``carrier`` untouched when tracing is disabled."""
+        return None
+
+    def extract_carrier(
+        self, carrier: "Mapping[str, str]"
+    ) -> "Optional[Mapping[str, str]]":
+        """No-op: no parent context to continue when tracing is disabled."""
+        return None
+
 
 # Shared singleton: the default no-op hook is stateless, so one instance is
 # reused everywhere a gateway needs a zero-cost default.
@@ -5517,6 +5662,174 @@ class _TurnLeaseHold:
             self._token = None
 
 
+# ---------------------------------------------------------------------------
+# Per-session turn-execution isolation (Issue #4011)
+#
+# Today every session's agent turn runs as an ``asyncio`` task on the single
+# gateway event loop, in one process: a turn that wedges the loop (a tight CPU
+# loop, a blocking C-extension call, GIL starvation), runs away, or crashes
+# the interpreter degrades or kills *every* session at once, and the only
+# remedy the runtime has is a whole-process ``os._exit(75)`` restart that
+# drops every other user's in-flight turn.
+#
+# This is the pure, dependency-free contract for a swappable turn *executor*,
+# mirroring how every other gateway robustness knob (drain, admission,
+# rate-limit, liveness, dead-letter, turn-lock) is a swappable protocol in
+# core. The executor decides *where* a session's turn runs; the default
+# :class:`InProcessTurnExecutor` runs it on the current loop exactly as today
+# (zero change for existing users, no dependency introduced). Heavy executors
+# that place a turn on an isolated worker (subprocess / container / remote),
+# each with optional per-worker resource limits, are optional-dep runtime
+# concerns that live in the wrapper/bot package; core keeps only the protocol,
+# the placement/fencing invariants, and the safe in-process default so a bad
+# turn can be torn down on its own worker without an ``os._exit`` of the whole
+# gateway.
+# ---------------------------------------------------------------------------
+
+
+class WorkerWedgedError(Exception):
+    """Raised when a session's turn worker is wedged / unresponsive.
+
+    Signals the gateway that the placement's worker can no longer make
+    progress (a wedged loop, an unresponsive subprocess/container, a lost
+    remote worker). The gateway responds by tearing down *only* that
+    placement (:meth:`TurnExecutorProtocol.teardown`) and re-placing the
+    session, rather than the process-wide ``os._exit`` that a single shared
+    loop forces today. Distinct from ordinary turn errors, which surface as
+    the turn's own exception and do not condemn the worker.
+    """
+
+
+@dataclass(frozen=True)
+class TurnPlacement:
+    """Which worker owns a session's turns, and at what epoch (Issue #4011).
+
+    A lifecycle-owned handle returned by :meth:`TurnExecutorProtocol.place`
+    and passed back to :meth:`~TurnExecutorProtocol.execute_turn` /
+    :meth:`~TurnExecutorProtocol.teardown`. The gateway revalidates the
+    placement (worker alive, matching ``epoch``) immediately before dispatch
+    and again after any awaited admission/approval work; a stale placement
+    (its worker was replaced, so ``epoch`` no longer matches) **fails closed**
+    and the session is re-placed rather than executed on a dead worker.
+
+    Attributes:
+        session_id: The resolved session whose turns this placement serves.
+        worker_id: Opaque id of the worker that owns the session's turns. For
+            the in-process default this is a constant (one loop, one worker);
+            for isolated executors it identifies the subprocess/container/
+            remote worker.
+        epoch: Monotonic generation bumped whenever the worker backing a
+            session is replaced (e.g. after a wedge teardown). A turn carrying
+            a stale epoch must not run — the fencing token that stops a
+            reclaimed worker from executing against a session it no longer owns.
+    """
+
+    session_id: str
+    worker_id: str
+    epoch: int = 0
+
+
+@runtime_checkable
+class TurnExecutorProtocol(Protocol):
+    """Contract for *where* a session's agent turn executes (Issue #4011).
+
+    The gateway calls :meth:`place` to obtain a :class:`TurnPlacement` for a
+    session, then :meth:`execute_turn` to run the turn on that placement, and
+    :meth:`teardown` to reclaim a placement's worker. The default
+    :class:`InProcessTurnExecutor` runs the turn on the current event loop,
+    reproducing today's behaviour exactly; an isolated executor (subprocess /
+    container / remote, in the wrapper) runs it on its own worker so a wedged,
+    runaway, or crashed turn is contained to its session.
+
+    Contract:
+        * :meth:`place` returns the placement that owns ``session_id``'s
+          turns. Implementations may cache a placement per session and bump
+          its ``epoch`` when the backing worker is replaced.
+        * :meth:`execute_turn` runs ``turn`` (an ``async`` no-arg callable the
+          gateway already built for this turn) on ``placement`` and returns
+          its result. The in-process default awaits ``turn`` directly on the
+          current loop. An isolated executor does **not** ship this live
+          callable (and its captured loop/agent state) across a process
+          boundary — the worker owns the session (via :meth:`place`) and
+          rebuilds/dispatches the turn from serialisable inputs on its own
+          side; ``turn`` then acts as the gateway-side await point for that
+          worker's result. ``cancel_token`` carries the per-turn interrupt so
+          the existing cancellation seam is preserved. ``limits`` optionally
+          bounds the worker's CPU/memory/wall time (honoured by isolated
+          executors; ignored in-process). It raises :class:`WorkerWedgedError`
+          if the placement's worker can no longer make progress.
+        * :meth:`teardown` reclaims ``placement``'s worker (kills the
+          subprocess/container/remote worker, or is a no-op in-process),
+          scoped to the owning session only.
+
+    A worker fault must fail *scoped*: tear down the offending placement and
+    re-place its session, never the whole gateway — mirroring the fail-safe,
+    blast-radius-contained defaults elsewhere.
+    """
+
+    async def place(self, session_id: str) -> TurnPlacement:
+        """Return the placement that owns ``session_id``'s turns."""
+        ...
+
+    async def execute_turn(
+        self,
+        placement: TurnPlacement,
+        turn: "Callable[[], Awaitable[Any]]",
+        *,
+        cancel_token: Any = None,
+        limits: Any = None,
+    ) -> Any:
+        """Run ``turn`` on ``placement`` and return its result."""
+        ...
+
+    async def teardown(self, placement: TurnPlacement, *, reason: str) -> None:
+        """Reclaim ``placement``'s worker (scoped to its session)."""
+        ...
+
+
+class InProcessTurnExecutor:
+    """Default turn executor — today's on-loop behaviour, no isolation.
+
+    Zero-cost, dependency-free implementation of
+    :class:`TurnExecutorProtocol` for single-process deployments. It runs each
+    turn directly on the current event loop exactly as the gateway does today,
+    so selecting it (or leaving the executor unset) is byte-for-byte backward
+    compatible and introduces no dependency. It provides **no** blast-radius
+    isolation: a turn that wedges the loop still affects the process — choosing
+    an isolated executor (subprocess / container / remote, in the wrapper) is
+    what contains a bad turn to its own worker.
+
+    All turns share one worker (the current process/loop), so every placement
+    carries the same constant ``worker_id`` at ``epoch`` 0, and ``teardown``
+    is inert — an in-process turn is cancelled through the existing
+    ``cancel_token`` when the holding task exits, so there is no worker to
+    reclaim. The ``limits`` argument is accepted for protocol symmetry but is
+    inert here (no separate worker to bound).
+    """
+
+    _WORKER_ID = "inprocess"
+
+    async def place(self, session_id: str) -> TurnPlacement:
+        return TurnPlacement(
+            session_id=session_id, worker_id=self._WORKER_ID, epoch=0
+        )
+
+    async def execute_turn(
+        self,
+        placement: TurnPlacement,
+        turn: "Callable[[], Awaitable[Any]]",
+        *,
+        cancel_token: Any = None,
+        limits: Any = None,
+    ) -> Any:
+        # Run the gateway's pre-built turn coroutine on this loop, unchanged.
+        return await turn()
+
+    async def teardown(self, placement: TurnPlacement, *, reason: str) -> None:
+        # No isolated worker to reclaim; the shared loop keeps serving.
+        return None
+
+
 # ---------------------------------------------------------------------------
 # Declarative method -> required-scope registry (Issue #3206)
 # ---------------------------------------------------------------------------

src/praisonai-agents/praisonaiagents/goal/judge.py

@@ -47,7 +47,11 @@ def _tail(text: str, limit: int = 4000) -> str:
     return text[-limit:]
 
 
-def _build_goal_judge_prompt(state: "GoalState", transcript_tail: str) -> str:
+def _build_goal_judge_prompt(
+    state: "GoalState",
+    transcript_tail: str,
+    verification_block: Optional[str] = None,
+) -> str:
     """Build a strict, criteria-anchored, JSON-only judge prompt."""
     lines = [
         "You are an independent completion judge for an autonomous agent.",
@@ -72,6 +76,13 @@ def _build_goal_judge_prompt(state: "GoalState", transcript_tail: str) -> str:
             lines.append(
                 "If ANY constraint is violated, you MUST answer 'continue'."
             )
+    if verification_block:
+        lines += [
+            "",
+            "EXECUTABLE VERIFICATION RESULTS (authoritative — command exit "
+            "codes and file checks; prefer this over prose claims):",
+            verification_block,
+        ]
     lines += [
         "",
         "AGENT OUTPUT (most recent):",
@@ -165,6 +176,7 @@ def judge_goal(
     transcript_tail: str,
     *,
     judge_model: Optional[str] = None,
+    verification_block: Optional[str] = None,
 ) -> Tuple[str, str]:
     """Judge whether ``state.goal`` is met given the latest agent output.
 
@@ -180,7 +192,7 @@ def judge_goal(
     """
     from ..eval.judge import Judge
 
-    prompt = _build_goal_judge_prompt(state, transcript_tail)
+    prompt = _build_goal_judge_prompt(state, transcript_tail, verification_block)
     try:
         judge = Judge(model=judge_model or _default_judge_model(), temperature=0.0)
         litellm = judge._get_litellm()

src/praisonai-agents/praisonaiagents/goal/loop.py

@@ -128,11 +128,44 @@ class GoalLoopMixin:
         if state is None or state.status != "active":
             return None
 
+        # Executable verification: run configured hooks and gate on the result.
+        # A failing BLOCKING hook short-circuits to "continue" WITHOUT spending
+        # a judge call. Passing/non-blocking results become authoritative
+        # evidence in the judge prompt.
+        verification_block: Optional[str] = None
+        if getattr(self, "_verification_hooks", None):
+            results = self._run_verification_hooks()
+            self._last_verification_results = results
+            self._record_verification_journal(results)
+            verification_block = self._verification_summary_tail(results)
+            blocking_failed = [
+                r for r in results
+                if r.get("blocking", True) and not r.get("success")
+            ]
+            if blocking_failed:
+                state.turns_used += 1
+                state.last_verdict = "continue"
+                # Carry the failing hooks' captured stdout/stderr into the
+                # continuation prompt so the next turn has the diagnostics
+                # needed to repair the failure (not just a bare label).
+                failure_detail = self._verification_summary_tail(blocking_failed)
+                reason = "verification gate failed"
+                if failure_detail:
+                    reason = f"{reason}:\n{failure_detail}"
+                state.last_reason = reason
+                if state.turns_used >= state.max_turns:
+                    state.status = "paused"
+                    self._persist_goal_state()
+                    return "budget_paused", reason
+                self._persist_goal_state()
+                return "continue", reason
+
         state.turns_used += 1
         verdict, reason = judge_goal(
             state,
             _tail(response),
             judge_model=getattr(self, "_goal_judge_model", None),
+            verification_block=verification_block,
         )
         state.last_verdict, state.last_reason = verdict, reason
 

src/praisonai-agents/praisonaiagents/guardrails/__init__.py

@@ -7,7 +7,12 @@ including both function-based and LLM-based guardrails.
 
 from .guardrail_result import GuardrailResult
 from .llm_guardrail import LLMGuardrail
-from .protocols import GuardrailProtocol, StructuralGuardrailProtocol, PolicyGuardrailProtocol
+from .protocols import (
+    GuardrailProtocol,
+    StructuralGuardrailProtocol,
+    PolicyGuardrailProtocol,
+    is_guardrail_object,
+)
 from .chain import GuardrailChain
 
 __all__ = [
@@ -16,5 +21,6 @@ __all__ = [
     "GuardrailProtocol", 
     "StructuralGuardrailProtocol", 
     "PolicyGuardrailProtocol", 
-    "GuardrailChain"
+    "GuardrailChain",
+    "is_guardrail_object",
 ]

src/praisonai-agents/praisonaiagents/guardrails/protocols.py

@@ -175,3 +175,25 @@ class PolicyGuardrailProtocol(Protocol):
         """
         ...
 
+
+# Method names that make an object a guardrail. An object exposing at least one
+# of these is a protocol-conforming guardrail even though it is not ``callable``.
+GUARDRAIL_PROTOCOL_METHODS = ("validate_input", "validate_output", "validate_tool_call")
+
+
+def is_guardrail_object(obj: Any) -> bool:
+    """Return True when ``obj`` is a protocol-conforming guardrail *instance*.
+
+    ``isinstance(obj, GuardrailProtocol)`` is not usable here: a ``runtime_checkable``
+    Protocol only checks method presence and would also match unrelated objects
+    partially, while classes (not instances) must be rejected. This helper is the
+    single place that decides "is this thing a guardrail object?".
+
+    Plain validator callables are handled by their own passthrough branch and are
+    not required to match here (they will, however, if they also expose one of the
+    protocol methods - e.g. ``GuardrailChain``).
+    """
+    if obj is None or isinstance(obj, type):
+        return False
+    return any(callable(getattr(obj, name, None)) for name in GUARDRAIL_PROTOCOL_METHODS)
+

src/praisonai-agents/praisonaiagents/hooks/__init__.py

@@ -104,6 +104,9 @@ __all__ = [
     # Verification hooks (protocols)
     "VerificationHook",
     "VerificationResult",
+    "BaseVerificationHook",
+    "CommandVerificationHook",
+    "FileCheckHook",
     # Simplified API (beginner-friendly)
     "add_hook",
     "remove_hook",
@@ -178,6 +181,9 @@ _LAZY_GROUPS = {
     'verification': {
         'VerificationHook': ('praisonaiagents.hooks.verification', 'VerificationHook'),
         'VerificationResult': ('praisonaiagents.hooks.verification', 'VerificationResult'),
+        'BaseVerificationHook': ('praisonaiagents.hooks.verification', 'BaseVerificationHook'),
+        'CommandVerificationHook': ('praisonaiagents.hooks.verification', 'CommandVerificationHook'),
+        'FileCheckHook': ('praisonaiagents.hooks.verification', 'FileCheckHook'),
     },
     # Simplified API (beginner-friendly aliases)
     'simplified_api': {

src/praisonai-agents/praisonaiagents/hooks/runner.py

@@ -184,7 +184,19 @@ class HookRunner:
             self._execute_single(hook, event, input_data)
             for hook in hooks
         ]
-        return await asyncio.gather(*tasks)
+        results = await asyncio.gather(*tasks)
+
+        # Parallel hooks cannot mutate input_data (only sequential hooks apply
+        # modified_input). Warn if a parallel hook returns modifications that
+        # will be silently discarded, so hook authors mark it sequential=True.
+        for hook, result in zip(hooks, results):
+            if result.success and result.output and result.output.modified_input:
+                logger.warning(
+                    "Parallel hook '%s' returned modified_input which is "
+                    "discarded; register it with sequential=True to apply it.",
+                    hook.name,
+                )
+        return results
     
     async def _execute_sequential(
         self,

src/praisonai-agents/praisonaiagents/hooks/verification.py

@@ -114,17 +114,26 @@ class BaseVerificationHook:
     
     name: str = "base"
     timeout_seconds: float = 60.0
-    
-    def __init__(self, name: Optional[str] = None, timeout: float = 60.0):
+    blocking: bool = True
+
+    def __init__(
+        self,
+        name: Optional[str] = None,
+        timeout: float = 60.0,
+        blocking: bool = True,
+    ):
         """Initialize the hook.
         
         Args:
             name: Override the hook name
             timeout: Timeout in seconds for execution
+            blocking: When True (default), a failure of this hook blocks the
+                autonomous loop from finalising with a success outcome.
         """
         if name:
             self.name = name
         self.timeout_seconds = timeout
+        self.blocking = blocking
     
     def run(self, context: Optional[Dict[str, Any]] = None) -> VerificationResult:
         """Run the verification hook.
@@ -177,19 +186,25 @@ class CommandVerificationHook(BaseVerificationHook):
     def __init__(
         self,
         name: str,
-        command: list,
+        command,
         cwd: Optional[str] = None,
         timeout: float = 60.0,
+        blocking: bool = True,
     ):
         """Initialize command hook.
         
         Args:
             name: Hook name
-            command: Command to run as list (e.g., ["pytest", "-v"])
+            command: Command to run as a list (e.g., ["pytest", "-v"]) or a
+                string (e.g., "python -m pytest -q"), which is shell-split.
             cwd: Working directory for command
             timeout: Timeout in seconds
+            blocking: When True (default), a failure blocks completion.
         """
-        super().__init__(name=name, timeout=timeout)
+        super().__init__(name=name, timeout=timeout, blocking=blocking)
+        if isinstance(command, str):
+            import shlex
+            command = shlex.split(command)
         self.command = command
         self.cwd = cwd
     
@@ -235,3 +250,143 @@ class CommandVerificationHook(BaseVerificationHook):
                 output=str(e),
                 error=str(e),
             )
+
+
+def _get_dotted(data: Any, path: str) -> Any:
+    """Access a dotted path (e.g. ``"a.b.c"``) in nested dicts/lists.
+
+    Returns ``None`` if any segment is missing. List indices are supported
+    via integer-looking segments.
+    """
+    value = data
+    for part in path.split("."):
+        if isinstance(value, dict):
+            value = value.get(part)
+        elif isinstance(value, (list, tuple)):
+            try:
+                value = value[int(part)]
+            except (ValueError, IndexError):
+                return None
+        else:
+            return None
+    return value
+
+
+class FileCheckHook(BaseVerificationHook):
+    """Declarative file-assertion verification hook (no shell involved).
+
+    Checks portable filesystem facts:
+      - ``exists``    : the file exists (default True).
+      - ``non_empty`` : the file has non-zero size.
+      - ``contains``  : the file text contains a substring.
+      - ``json_field``/``equals`` : the JSON value at a dotted path equals a value.
+
+    Example:
+        hook = FileCheckHook(name="changelog", path="CHANGELOG.md", non_empty=True)
+        hook = FileCheckHook(name="cfg", path="out.json",
+                             json_field="status.code", equals=0)
+    """
+
+    def __init__(
+        self,
+        name: str,
+        path: str,
+        exists: bool = True,
+        non_empty: bool = False,
+        contains: Optional[str] = None,
+        json_field: Optional[str] = None,
+        equals: Any = None,
+        blocking: bool = True,
+    ):
+        super().__init__(name=name, blocking=blocking)
+        self.path = path
+        self.exists = exists
+        self.non_empty = non_empty
+        self.contains = contains
+        self.json_field = json_field
+        self.equals = equals
+
+    def _execute(self, context: Optional[Dict[str, Any]] = None) -> VerificationResult:
+        import os
+
+        details: Dict[str, Any] = {"path": self.path}
+        file_exists = os.path.isfile(self.path)
+        details["exists"] = file_exists
+
+        if self.exists and not file_exists:
+            return VerificationResult(
+                success=False,
+                output=f"File not found: {self.path}",
+                error="not_found",
+                details=details,
+            )
+        if not self.exists:
+            success = not file_exists
+            return VerificationResult(
+                success=success,
+                output=(
+                    f"File exists but was expected absent: {self.path}"
+                    if not success else f"File absent as expected: {self.path}"
+                ),
+                details=details,
+            )
+
+        if self.non_empty:
+            size = os.path.getsize(self.path)
+            details["size"] = size
+            if size == 0:
+                return VerificationResult(
+                    success=False,
+                    output=f"File is empty: {self.path}",
+                    error="empty",
+                    details=details,
+                )
+
+        if self.contains is not None or self.json_field is not None:
+            try:
+                with open(self.path, "r", encoding="utf-8") as fh:
+                    text = fh.read()
+            except Exception as e:
+                return VerificationResult(
+                    success=False, output=str(e), error=str(e), details=details
+                )
+
+            if self.contains is not None and self.contains not in text:
+                return VerificationResult(
+                    success=False,
+                    output=f"File {self.path} does not contain {self.contains!r}",
+                    error="missing_substring",
+                    details=details,
+                )
+
+            if self.json_field is not None:
+                import json as _json
+
+                try:
+                    data = _json.loads(text)
+                except Exception as e:
+                    return VerificationResult(
+                        success=False,
+                        output=f"File {self.path} is not valid JSON: {e}",
+                        error="invalid_json",
+                        details=details,
+                    )
+                actual = _get_dotted(data, self.json_field)
+                details["json_field"] = self.json_field
+                details["actual"] = actual
+                if actual != self.equals:
+                    return VerificationResult(
+                        success=False,
+                        output=(
+                            f"{self.path}:{self.json_field} = {actual!r}, "
+                            f"expected {self.equals!r}"
+                        ),
+                        error="mismatch",
+                        details=details,
+                    )
+
+        return VerificationResult(
+            success=True,
+            output=f"File check passed: {self.path}",
+            details=details,
+        )

src/praisonai-agents/praisonaiagents/knowledge/chunking.py

@@ -2,6 +2,30 @@ from typing import List, Union, Optional, Dict, Any
 from functools import cached_property
 import importlib
 
+#: Names accepted for a chunking strategy that are not chunker ids themselves.
+#: ``ChunkingStrategy.FIXED`` and ``.PARAGRAPH`` had no chunker at all, so they
+#: silently did nothing; they now map onto the chunkers that implement them.
+CHUNKER_ALIASES = {
+    "fixed": "token",
+    "paragraph": "recursive",
+}
+
+
+def normalize_chunker_type(name: str) -> str:
+    """Map a user-facing chunking strategy name onto a chunker id."""
+    if not name:
+        return "recursive"
+    key = str(name).strip().lower()
+    key = CHUNKER_ALIASES.get(key, key)
+    if key not in Chunking.CHUNKER_PARAMS:
+        from ..config.parse_utils import make_preset_error
+        raise make_preset_error(
+            "chunking strategy", name,
+            sorted(set(Chunking.CHUNKER_PARAMS) | set(CHUNKER_ALIASES)),
+        )
+    return key
+
+
 class Chunking:
     """A unified class for text chunking with various chunking strategies."""
     

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

@@ -2,7 +2,7 @@ import os
 import logging
 from praisonaiagents._logging import get_logger
 from datetime import datetime
-from .chunking import Chunking
+from .chunking import Chunking, normalize_chunker_type
 from functools import cached_property
 
 
@@ -162,6 +162,10 @@ class Knowledge:
             # Merge graph_store config if provided (for graph memory support)
             if "graph_store" in self._config:
                 base_config["graph_store"] = self._config["graph_store"]
+
+            # Merge chunker config if provided (consumed by self.chunker)
+            if "chunker" in self._config:
+                base_config["chunker"] = self._config["chunker"]
         return base_config
 
     def _prepare_mem0_config(self, config):
@@ -195,28 +199,46 @@ class Knowledge:
     def memory(self):
         """Initialize knowledge adapter using protocol-driven approach."""
         # Import registry functions
-        from .adapters import get_knowledge_adapter, get_first_available_knowledge_adapter
+        from .adapters import (
+            get_knowledge_adapter,
+            get_first_available_knowledge_adapter,
+            has_knowledge_adapter,
+            list_knowledge_adapters,
+        )
         
         # Determine provider preference. Distinguish an explicit user choice from
         # the implicit default so fallback diagnostics are only loud when a
         # configured backend degrades (see issue #2972 / PR #2982 review).
         # ``self.config`` is always merged with defaults, so use the raw
         # user-supplied ``_config`` to detect whether the provider was chosen.
-        provider_explicit = bool(
-            (self._config or {}).get("vector_store", {}).get("provider")
+        # Presence of the key (not truthiness) marks an explicit choice, so an
+        # explicitly empty/null provider is treated as a misconfiguration and
+        # surfaces the preset error instead of silently degrading to Mem0.
+        provider_explicit = "provider" in (
+            (self._config or {}).get("vector_store", {}) or {}
         )
         provider = self.config.get("vector_store", {}).get("provider", "mem0")
         self._log(f"Requested knowledge provider: {provider}")
         
-        # Map legacy provider names to adapter names
+        # Legacy provider aliases. Anything else is looked up in the adapter
+        # registry, so adapters added with register_knowledge_adapter() are
+        # reachable; only genuinely unknown names fall back (and say so).
         provider_mapping = {
-            "chroma": "chroma",
-            "mongodb": "mongodb", 
-            "mem0": "mem0",
-            "sqlite": "sqlite"
+            "chromadb": "chroma",
+            "rag": "chroma",
         }
-        
-        adapter_name = provider_mapping.get(provider, "mem0")
+
+        adapter_name = provider_mapping.get(provider, provider)
+        if not has_knowledge_adapter(adapter_name):
+            from ..config.parse_utils import make_preset_error
+            err = make_preset_error(
+                "knowledge vector_store provider", provider,
+                sorted(set(list_knowledge_adapters()) | set(provider_mapping)),
+            )
+            if provider_explicit:
+                raise err
+            self._log(f"{err} Falling back to 'mem0'.")
+            adapter_name = "mem0"
         
         # Try to get preferred adapter, fallback to available ones
         try:
@@ -287,10 +309,13 @@ class Knowledge:
 
     @cached_property
     def chunker(self):
+        # Read from config; the previous hardcoded recursive/512/50 made
+        # KnowledgeConfig.chunk_size / chunking_strategy / chunker inert.
+        chunker_cfg = self.config.get("chunker") or {}
         return Chunking(
-            chunker_type='recursive',
-            chunk_size=512,
-            chunk_overlap=50
+            chunker_type=normalize_chunker_type(chunker_cfg.get("type", "recursive")),
+            chunk_size=chunker_cfg.get("chunk_size", 512),
+            chunk_overlap=chunker_cfg.get("chunk_overlap", 50),
         )
 
     def _log(self, message, level=2):
@@ -413,7 +438,11 @@ class Knowledge:
 
     def history(self, memory_id):
         """Get the history of changes for a memory."""
-        return self.memory.history(memory_id)
+        if hasattr(self.memory, "history"):
+            return self.memory.history(memory_id)
+        raise NotImplementedError(
+            f"{type(self.memory).__name__} does not support history()"
+        )
 
     def delete(self, memory_id):
         """Delete a memory."""
@@ -421,11 +450,21 @@ class Knowledge:
 
     def delete_all(self, user_id=None, agent_id=None, run_id=None):
         """Delete all memories."""
-        self.memory.delete_all(user_id=user_id, agent_id=agent_id, run_id=run_id)
+        if hasattr(self.memory, "delete_all"):
+            self.memory.delete_all(user_id=user_id, agent_id=agent_id, run_id=run_id)
+        else:
+            raise NotImplementedError(
+                f"{type(self.memory).__name__} does not support delete_all()"
+            )
 
     def reset(self):
         """Reset all memories."""
-        self.memory.reset()
+        if hasattr(self.memory, "reset"):
+            self.memory.reset()
+        else:
+            logger.warning(
+                f"{type(self.memory).__name__} does not support reset(); no-op"
+            )
 
     def normalize_content(self, content):
         """Normalize content for consistent storage."""

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

@@ -534,7 +534,18 @@ Respond with ONLY a valid JSON tool call in this format:
         if self._failover_manager:
             self._current_profile = self._failover_manager.get_next_profile()
             if self._current_profile:
-                self._switch_to_profile(self._current_profile)
+                # Apply the initial profile's values to the instance attributes
+                # (backward compatible with the documented llm.api_key/model
+                # surface). Later failover rotations are per-call only and do
+                # not mutate these attributes (issue #3613 gap 2).
+                profile = self._current_profile
+                if profile.api_key:
+                    self.api_key = profile.api_key
+                if profile.base_url:
+                    self.base_url = profile.base_url
+                if profile.model and profile.model != self.model:
+                    logging.info(f"Failover: using profile model {profile.model}")
+                    self.model = profile.model
 
         # Cache for formatted tools and messages
         self._formatted_tools_cache = {}
@@ -974,18 +985,58 @@ Respond with ONLY a valid JSON tool call in this format:
 
     def _switch_to_profile(self, profile: "AuthProfile") -> None:
         """Switch to a new auth profile for failover.
-        
+
+        The profile's credentials are threaded through the per-call kwargs of
+        the retry loop instead of being written onto the shared instance. The
+        old behaviour mutated ``self.api_key``/``self.base_url``/``self.model``
+        on a shared ``LLM`` (issue #3613 gap 2), so concurrent requests on one
+        instance could silently cross-talk credentials and model after a
+        failover rotation. Keeping the instance attributes on the initial
+        profile preserves single-profile behaviour and the documented
+        ``llm.api_key``/``llm.model`` surface.
+
         Args:
             profile: AuthProfile to switch to
         """
+        # NOTE: profile values are applied to the per-call kwargs by
+        # _handle_retry_exception when it rotates. This method is kept as a
+        # no-op placeholder so any out-of-tree callers that invoke it directly
+        # keep working; it intentionally no longer mutates shared state.
+        if profile is None:
+            return
+
+    def _response_model_for_tracking(self, response: Any) -> str:
+        """Model to attribute token usage to for a completed call.
+
+        Prefers the model carried on the response itself, which reflects the
+        provider/profile that actually served the request. This is race-free
+        under concurrency (unlike reading the shared ``self._current_profile``)
+        and stays correct after a failover rotation. Falls back to
+        ``self.model`` when the response does not expose a model.
+        """
+        model = None
+        if isinstance(response, dict):
+            model = response.get("model")
+        else:
+            model = getattr(response, "model", None)
+        return model or self.model
+
+    def _apply_profile_to_kwargs(self, profile: "AuthProfile", kwargs: dict) -> dict:
+        """Return a new kwargs dict with profile overrides applied.
+
+        Never mutates ``self``: the active profile is a per-call concern, so
+        its credentials travel in the request kwargs only. Callers that pass
+        an explicit per-call override (e.g. ``api_key=``) keep it; profile
+        values fill in the gaps.
+        """
+        kwargs = dict(kwargs)
         if profile.api_key:
-            self.api_key = profile.api_key
+            kwargs["api_key"] = profile.api_key
         if profile.base_url:
-            self.base_url = profile.base_url
-        if profile.model and profile.model != self.model:
-            # Only log if model actually changes
-            logging.info(f"Failover: switching from {self.model} to {profile.model}")
-            self.model = profile.model
+            kwargs["base_url"] = profile.base_url
+        if profile.model:
+            kwargs["model"] = profile.model
+        return kwargs
 
     def _resolve_subscription_creds(self):
         """Lazy resolve + cache subscription credentials, checking expiration."""
@@ -1055,7 +1106,7 @@ Respond with ONLY a valid JSON tool call in this format:
             )
             return None
 
-    def _handle_retry_exception(self, e, attempt, kwargs, refreshed_creds=None):
+    def _handle_retry_exception(self, e, attempt, kwargs, refreshed_creds=None, active_profile=None):
         """Decide retry vs raise from a failover decision. No sleep/await.
 
         Shared decision logic used by both the sync (`_call_with_retry`) and
@@ -1069,21 +1120,34 @@ Respond with ONLY a valid JSON tool call in this format:
         blocks the event loop. The already-refreshed credentials (if any) are
         passed in via ``refreshed_creds``.
 
+        The active failover profile is a *per-call* value threaded in via
+        ``active_profile`` and returned back in the result tuple. It is never
+        read from or written to ``self._current_profile`` here, so concurrent
+        requests sharing one ``LLM`` instance cannot corrupt each other's
+        failover bookkeeping or cost attribution (issue #3613 gap 2).
+
         Args:
             e: The exception raised by the wrapped call.
             attempt: Zero-based attempt index for the current iteration.
             kwargs: Keyword arguments for the wrapped call (may be updated).
             refreshed_creds: Credentials already refreshed by the caller (off
                 the event loop for async), or ``None`` if no refresh was done.
+            active_profile: The profile this call is currently using, or
+                ``None``. Defaults to the initial profile for backward compat.
 
         Returns:
-            Tuple of (should_raise, category, retry_delay, error_str, kwargs):
+            Tuple of (should_raise, category, retry_delay, error_str, kwargs,
+            active_profile):
                 should_raise: Whether the caller should re-raise immediately.
                 category: The error category/reason from the decision.
                 retry_delay: Seconds to wait before the next attempt.
                 error_str: String form of the exception.
                 kwargs: The (possibly updated) keyword arguments.
+                active_profile: The (possibly rotated) profile for the next
+                    attempt.
         """
+        if active_profile is None:
+            active_profile = self._current_profile
         # A turn is "side-effecting" when it exposes tools the model may call:
         # a post-dispatch failure on such a turn could have already run a
         # side-effecting tool, so it must not be silently replayed. A turn with
@@ -1135,55 +1199,48 @@ Respond with ONLY a valid JSON tool call in this format:
         # immediately with fresh credentials and skip profile rotation so we
         # neither mark the current profile failed nor overwrite the new creds.
         if refreshed_auth:
-            return False, category, retry_delay, error_str, kwargs
+            return False, category, retry_delay, error_str, kwargs, active_profile
 
-        # Handle different failover decision actions
+        # Handle different failover decision actions. Rotation is a per-call
+        # concern: the active profile is a local threaded in/out of this method,
+        # never self._current_profile, so concurrent calls on one shared LLM
+        # cannot mark each other's profiles failed/healthy (issue #3613 gap 2).
         if decision.action == "rotate_profile" and self._failover_manager:
-            if self._current_profile:
+            if active_profile:
                 is_rate_limit = (category == "rate_limit")
                 self._failover_manager.mark_failure(
-                    self._current_profile, error_str, is_rate_limit=is_rate_limit
+                    active_profile, error_str, is_rate_limit=is_rate_limit
                 )
             next_profile = self._failover_manager.get_next_profile()
-            if next_profile and next_profile != self._current_profile:
-                self._switch_to_profile(next_profile)
-                self._current_profile = next_profile
-                # Update the kwargs with new profile values for the next retry
-                if "api_key" in kwargs:
-                    kwargs["api_key"] = self.api_key
-                if "base_url" in kwargs:
-                    kwargs["base_url"] = self.base_url
-                if "model" in kwargs:
-                    kwargs["model"] = self.model
+            if next_profile and next_profile != active_profile:
+                active_profile = next_profile
+                # Thread the active profile through the per-call kwargs so the
+                # shared instance state is never mutated (issue #3613 gap 2).
+                kwargs = self._apply_profile_to_kwargs(next_profile, kwargs)
                 logging.info(f"Failover: switched to profile '{next_profile.name}'")
             else:
                 # No alternate profile available - retrying would just hit the
                 # same failing profile, so surface the error instead.
                 can_retry = False
         # Legacy failover for compatibility (when decision is retry but failover is configured)
-        elif self._failover_manager and self._current_profile and decision.action == "retry":
+        elif self._failover_manager and active_profile and decision.action == "retry":
             is_rate_limit = (category == "rate_limit")
             self._failover_manager.mark_failure(
-                self._current_profile, error_str, is_rate_limit=is_rate_limit
+                active_profile, error_str, is_rate_limit=is_rate_limit
             )
             next_profile = self._failover_manager.get_next_profile()
-            if next_profile and next_profile != self._current_profile:
-                self._switch_to_profile(next_profile)
-                self._current_profile = next_profile
-                # Update the kwargs with new profile values for the next retry
-                if "api_key" in kwargs:
-                    kwargs["api_key"] = self.api_key
-                if "base_url" in kwargs:
-                    kwargs["base_url"] = self.base_url
-                if "model" in kwargs:
-                    kwargs["model"] = self.model
+            if next_profile and next_profile != active_profile:
+                active_profile = next_profile
+                # Thread the active profile through the per-call kwargs so the
+                # shared instance state is never mutated (issue #3613 gap 2).
+                kwargs = self._apply_profile_to_kwargs(next_profile, kwargs)
                 # Enable retry for profile switch even if originally non-retryable
                 can_retry = True
                 retry_delay = 0.0
                 logging.info(f"Failover: switched to profile '{next_profile.name}'")
 
         should_raise = decision.action == "surface_error" or not can_retry
-        return should_raise, category, retry_delay, error_str, kwargs
+        return should_raise, category, retry_delay, error_str, kwargs, active_profile
 
     def _call_with_retry(self, func, *args, **kwargs):
         """Call a function with automatic retry on rate limit errors and failover support.
@@ -1200,6 +1257,10 @@ Respond with ONLY a valid JSON tool call in this format:
             The original exception if max retries exceeded
         """
         last_error = None
+        # Track the failover profile per call (never self._current_profile) so
+        # concurrent calls on one shared LLM cannot corrupt each other's
+        # failover bookkeeping (issue #3613 gap 2).
+        active_profile = self._current_profile
 
         for attempt in range(self._max_retries + 1):
             try:
@@ -1210,8 +1271,8 @@ Respond with ONLY a valid JSON tool call in this format:
                 result = func(*args, **kwargs)
                 
                 # Mark success if failover is configured
-                if self._failover_manager and self._current_profile:
-                    self._failover_manager.mark_success(self._current_profile)
+                if self._failover_manager and active_profile:
+                    self._failover_manager.mark_success(active_profile)
                 
                 # Reset idle timeout circuit breaker on success
                 self._idle_timeout_breaker.reset()
@@ -1226,8 +1287,12 @@ Respond with ONLY a valid JSON tool call in this format:
                 if self._should_attempt_auth_refresh(e, attempt):
                     refreshed_creds = self._try_refresh_subscription_creds()
 
-                should_raise, category, retry_delay, error_str, kwargs = (
-                    self._handle_retry_exception(e, attempt, kwargs, refreshed_creds=refreshed_creds)
+                should_raise, category, retry_delay, error_str, kwargs, active_profile = (
+                    self._handle_retry_exception(
+                        e, attempt, kwargs,
+                        refreshed_creds=refreshed_creds,
+                        active_profile=active_profile,
+                    )
                 )
 
                 if should_raise:
@@ -1276,6 +1341,10 @@ Respond with ONLY a valid JSON tool call in this format:
             The original exception if max retries exceeded
         """
         last_error = None
+        # Track the failover profile per call (never self._current_profile) so
+        # concurrent coroutines on one shared LLM cannot corrupt each other's
+        # failover bookkeeping (issue #3613 gap 2).
+        active_profile = self._current_profile
 
         for attempt in range(self._max_retries + 1):
             try:
@@ -1286,8 +1355,8 @@ Respond with ONLY a valid JSON tool call in this format:
                 result = await func(*args, **kwargs)
                 
                 # Mark success if failover is configured
-                if self._failover_manager and self._current_profile:
-                    self._failover_manager.mark_success(self._current_profile)
+                if self._failover_manager and active_profile:
+                    self._failover_manager.mark_success(active_profile)
                 
                 # Reset idle timeout circuit breaker on success
                 self._idle_timeout_breaker.reset()
@@ -1305,8 +1374,12 @@ Respond with ONLY a valid JSON tool call in this format:
                         self._try_refresh_subscription_creds
                     )
 
-                should_raise, category, retry_delay, error_str, kwargs = (
-                    self._handle_retry_exception(e, attempt, kwargs, refreshed_creds=refreshed_creds)
+                should_raise, category, retry_delay, error_str, kwargs, active_profile = (
+                    self._handle_retry_exception(
+                        e, attempt, kwargs,
+                        refreshed_creds=refreshed_creds,
+                        active_profile=active_profile,
+                    )
                 )
 
                 if should_raise:
@@ -1356,7 +1429,10 @@ Respond with ONLY a valid JSON tool call in this format:
             The completion response from litellm
         """
         import litellm
-        return self._call_with_retry(litellm.completion, **completion_params)
+        response = self._call_with_retry(litellm.completion, **completion_params)
+        if not completion_params.get("stream"):
+            self._track_token_usage(response, self._response_model_for_tracking(response))
+        return response
 
     async def _acompletion_with_retry(self, **completion_params):
         """Execute litellm.acompletion with automatic retry on rate limit errors.
@@ -1371,7 +1447,13 @@ Respond with ONLY a valid JSON tool call in this format:
             The completion response from litellm
         """
         import litellm
-        return await self._call_with_retry_async(litellm.acompletion, **completion_params)
+        response = await self._call_with_retry_async(
+            litellm.acompletion,
+            **completion_params,
+        )
+        if not completion_params.get("stream"):
+            self._track_token_usage(response, self._response_model_for_tracking(response))
+        return response
 
     def _supports_web_search(self) -> bool:
         """
@@ -1757,7 +1839,11 @@ Respond with ONLY a valid JSON tool call in this format:
                 
         except (KeyError, json.JSONDecodeError, TypeError) as e:
             logging.error(f"Error parsing tool call arguments: {e}")
-            function_name = tool_call.get("name", "unknown_function")
+            function_name = (
+                tool_call.get("function", {}).get("name")
+                if isinstance(tool_call.get("function"), dict)
+                else None
+            ) or tool_call.get("name", "unknown_function")
             arguments = {}
             tool_call_id = tool_call.get("id", f"tool_{id(tool_call)}")
             
@@ -2646,7 +2732,11 @@ Respond with ONLY a valid JSON tool call in this format:
                             resp = self._call_responses_api(**responses_params)
                             response_text, tool_calls, _reasoning = self._extract_from_responses_output(resp)
 
-                            # Track token usage
+                            # Usage accounting is independent from metrics display.
+                            # Quiet/default agents must still update the public
+                            # token collector after a paid model call.
+                            self._track_token_usage(resp, self.model)
+
                             if self.metrics and hasattr(resp, 'usage') and resp.usage:
                                 usage = resp.usage
                                 tokens_in = getattr(usage, 'input_tokens', 0) or 0
@@ -2869,10 +2959,6 @@ Respond with ONLY a valid JSON tool call in this format:
                                 is_reasoning=False
                             ))
                         
-                        # Track token usage
-                        if self.metrics:
-                            self._track_token_usage(final_response, self.model)
-                        
                         # Trigger llm_end callback with metrics for debug output
                         llm_latency_ms = (time.time() - current_time) * 1000
                         
@@ -3086,9 +3172,6 @@ Respond with ONLY a valid JSON tool call in this format:
                                             response_content = final_response["choices"][0]["message"].get("content")
                                             response_text = response_content if response_content is not None else ""
                                             
-                                            # Track token usage
-                                            if self.metrics:
-                                                self._track_token_usage(final_response, self.model)
                                         
                                         # Execute callbacks and display based on verbose setting
                                         if verbose and not interaction_displayed:
@@ -3278,9 +3361,6 @@ Respond with ONLY a valid JSON tool call in this format:
                                 response_content = final_response["choices"][0]["message"].get("content")
                                 response_text = response_content if response_content is not None else ""
                                 
-                                # Track token usage
-                                if self.metrics:
-                                    self._track_token_usage(final_response, self.model)
                             
                             # Execute callbacks and display based on verbose setting
                             if verbose and not interaction_displayed:
@@ -5414,27 +5494,66 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                 litellm.callbacks.append(event)
         self._registered_callbacks = list(events)
 
-    def _track_token_usage(self, response: Dict[str, Any], model: str) -> Optional[TokenMetrics]:
+    def _track_token_usage(self, response: Any, model: str) -> Optional[TokenMetrics]:
         """Extract and track token usage from LLM response."""
         if not TokenMetrics or not get_token_collector:
             return None
-        
-        # Note: metrics check moved to call sites for performance
-        # This method should only be called when self.metrics=True
-        
+
         try:
-            usage = response.get("usage", {})
+            if isinstance(response, dict):
+                usage = response.get("usage", {})
+            else:
+                usage = getattr(response, "usage", None)
             if not usage:
                 return None
-            
+
+            def _usage_value(*names: str) -> int:
+                for name in names:
+                    if isinstance(usage, dict):
+                        value = usage.get(name)
+                    else:
+                        value = getattr(usage, name, None)
+                    if value is not None:
+                        return int(value or 0)
+                return 0
+
+            def _detail_value(detail_names: tuple[str, ...], name: str) -> int:
+                for detail_name in detail_names:
+                    details = (
+                        usage.get(detail_name)
+                        if isinstance(usage, dict)
+                        else getattr(usage, detail_name, None)
+                    )
+                    if details:
+                        value = (
+                            details.get(name)
+                            if isinstance(details, dict)
+                            else getattr(details, name, None)
+                        )
+                        if value is not None:
+                            return int(value or 0)
+                return 0
+
             # Extract token counts
             metrics = TokenMetrics(
-                input_tokens=usage.get("prompt_tokens", 0),
-                output_tokens=usage.get("completion_tokens", 0),
-                cached_tokens=usage.get("cached_tokens", 0),
-                reasoning_tokens=usage.get("reasoning_tokens", 0),
-                audio_input_tokens=usage.get("audio_input_tokens", 0),
-                audio_output_tokens=usage.get("audio_output_tokens", 0)
+                input_tokens=_usage_value("prompt_tokens", "input_tokens"),
+                output_tokens=_usage_value("completion_tokens", "output_tokens"),
+                cached_tokens=_usage_value("cached_tokens") or _detail_value(
+                    ("prompt_tokens_details", "input_tokens_details"),
+                    "cached_tokens",
+                ),
+                reasoning_tokens=_usage_value("reasoning_tokens") or _detail_value(
+                    ("completion_tokens_details", "output_tokens_details"),
+                    "reasoning_tokens",
+                ),
+                audio_input_tokens=_usage_value("audio_input_tokens") or _detail_value(
+                    ("prompt_tokens_details", "input_tokens_details"),
+                    "audio_tokens",
+                ),
+                audio_output_tokens=_usage_value("audio_output_tokens") or _detail_value(
+                    ("completion_tokens_details", "output_tokens_details"),
+                    "audio_tokens",
+                ),
             )
             
             # Store metrics
@@ -5845,7 +5964,16 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                         "output": msg.get("content", ""),
                     })
                 else:
-                    input_items.append(msg)
+                    # Translate Chat Completions multimodal content parts
+                    # (type=text / image_url) into Responses API parts
+                    # (input_text / input_image); local image paths become
+                    # data URLs. Plain-string content is passed through.
+                    from .openai_client import OpenAIClient
+                    item = dict(msg)
+                    item["content"] = OpenAIClient._build_responses_content(
+                        msg.get("content", "")
+                    )
+                    input_items.append(item)
 
         if instructions:
             params["instructions"] = instructions
@@ -5893,6 +6021,19 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
         if self.timeout:
             params["timeout"] = self.timeout
 
+        # ── Subscription credentials ────────────────────────────────────
+        # Mirror _build_completion_params: without this the Responses API
+        # transport silently ignores auth= and bills the plain API key.
+        creds = self._resolve_subscription_creds()
+        if creds:
+            params["api_key"] = creds.api_key
+            if creds.base_url:
+                params["base_url"] = creds.base_url
+            if creds.headers:
+                extra_headers = dict(params.get("extra_headers") or {})
+                extra_headers.update(creds.headers)
+                params["extra_headers"] = extra_headers
+
         # ── Structured output ───────────────────────────────────────────
         output_json = kwargs.pop("output_json", None)
         output_pydantic = kwargs.pop("output_pydantic", None)
@@ -5942,7 +6083,9 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
         Returns a ``ResponsesAPIResponse`` object.
         """
         import litellm
-        return await self._call_with_retry_async(litellm.aresponses, **params)
+        response = await self._call_with_retry_async(litellm.aresponses, **params)
+        self._track_token_usage(response, self._response_model_for_tracking(response))
+        return response
 
     def _extract_from_responses_output(self, response) -> tuple:
         """
@@ -6491,6 +6634,23 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
         if callable(function_or_name):
             # Function object passed directly
             func = function_or_name
+            # If it's a BaseTool/FunctionTool instance, honour its declared schema
+            # (respects @tool name= override and excludes Injected[T] params) instead
+            # of re-deriving from the raw wrapped function's signature.
+            if hasattr(func, 'get_schema'):
+                tool_def = func.get_schema()
+                if (
+                    isinstance(tool_def, dict)
+                    and isinstance(tool_def.get("function"), dict)
+                    and isinstance(tool_def["function"].get("parameters"), dict)
+                ):
+                    tool_def = tool_def.copy()
+                    tool_def["function"] = tool_def["function"].copy()
+                    tool_def["function"]["parameters"] = self._fix_array_schemas(
+                        tool_def["function"]["parameters"]
+                    )
+                logging.debug(f"Generated tool definition from get_schema(): {tool_def}")
+                return tool_def
             function_name = func.__name__
             logging.debug(f"Generating tool definition for callable: {function_name}")
         else:

src/praisonai-agents/praisonaiagents/llm/openai_client.py

@@ -6,7 +6,9 @@ supporting both synchronous and asynchronous operations.
 """
 
 import os
+import base64
 import logging
+import mimetypes
 from praisonaiagents._logging import get_logger
 import time
 import json
@@ -16,6 +18,7 @@ from typing import Any, Dict, List, Optional, Union, AsyncIterator, Iterator, Ca
 from pydantic import BaseModel
 from dataclasses import dataclass
 import inspect
+from pathlib import Path
 
 from ..errors import ToolExecutionError
 
@@ -34,6 +37,54 @@ def _durable_iteration_kwargs(execute_tool_fn: Callable, index: int) -> Dict[str
         return {"_durable_iteration_index": index}
     return {}
 
+
+def _handle_native_deferred_result(
+    tool_result: Any,
+    messages: List[Dict],
+    tool_call_id: Optional[str],
+    function_name: str,
+    history_sink: Optional[List[Dict]] = None,
+) -> Any:
+    """Register a directly-returned ``DeferredToolResult`` on the native path.
+
+    Parity with the LiteLLM loop (``llm.py:_register_deferred_if_any``): a tool
+    may return a ``DeferredToolResult`` to signal "started; will resolve later".
+    On the native OpenAI-SDK loop the value was previously stringified and the
+    handle lost, so the eventual background result was never re-injected. Here
+    we register the handle on the shared resolver so a later
+    ``resolve_deferred(handle_id, value)`` appends a follow-up tool message that
+    a subsequent turn will replay, and surface the ``note`` to the model now.
+
+    The re-injection target is ``history_sink`` when provided — this must be the
+    caller's *durable* conversation history (e.g. the agent's ``chat_history``),
+    since the background job typically completes after this native loop has
+    already returned and its per-call ``messages`` copy has been discarded. When
+    no durable sink is given we fall back to ``messages`` (correct only if the
+    handle resolves while the loop is still open), preserving prior behaviour.
+
+    Returns the value to record for this turn (the deferred ``note`` string when
+    deferred, otherwise the unchanged ``tool_result``). Never raises.
+    """
+    from ..tools.call_executor import DeferredToolResult
+    if not isinstance(tool_result, DeferredToolResult):
+        return tool_result
+    try:
+        from ..tools.call_executor import get_deferred_resolver
+        resolver = get_deferred_resolver()
+        sink = history_sink if history_sink is not None else messages
+
+        def _reinject(handle_id: str, value: Any, session_id: Optional[str]) -> None:
+            sink.append({
+                "role": "tool",
+                "tool_call_id": tool_call_id,
+                "content": f"[deferred:{function_name}] {value}",
+            })
+
+        resolver.register_if_absent(tool_result.handle_id, _reinject)
+    except Exception as e:  # noqa: BLE001 — registration must never break the turn
+        logging.debug("Native deferred registration skipped (non-fatal): %s", e)
+    return tool_result.note
+
 # Lazy imports for optional dependencies
 _openai_module = None
 _rich_console = None
@@ -149,6 +200,26 @@ class ToolCall:
     type: str
     function: Dict[str, Any]
 
+def _extract_tool_call_name_and_args(tool_call):
+    """Parse a tool call's function name and JSON arguments.
+
+    Handles both the ToolCall dataclass and OpenAI-SDK objects. Returns a
+    tuple of (name, args, tool_call_id, err); err is None on success and the
+    caught exception otherwise. Callers keep their own failure handling.
+    """
+    tool_call_id = tool_call.id if hasattr(tool_call, 'id') else tool_call.get('id')
+    function_name = None
+    try:
+        if isinstance(tool_call, ToolCall):
+            function_name = tool_call.function["name"]
+            raw_arguments = tool_call.function["arguments"]
+        else:
+            function_name = tool_call.function.name
+            raw_arguments = tool_call.function.arguments
+        return function_name, json.loads(raw_arguments), tool_call_id, None
+    except (json.JSONDecodeError, KeyError, AttributeError, TypeError) as e:
+        return function_name, None, tool_call_id, e
+
 def process_stream_chunks(chunks):
     """Process streaming chunks into combined response"""
     if not chunks:
@@ -326,6 +397,11 @@ class OpenAIClient:
         self._formatted_tools_cache = {}
         self._fixed_schema_cache = {}
         self._max_cache_size = 100
+
+        # Token usage from the most recent completion (Issue #3933). Populated
+        # regardless of any display/metrics flag so headless callers and CLI
+        # envelopes can read real spend after a successful run.
+        self.last_token_metrics = None
     
     @property
     def console(self):
@@ -675,7 +751,11 @@ class OpenAIClient:
                         "output": msg.get("content", ""),
                     })
                 else:
-                    input_items.append(msg)
+                    item = dict(msg)
+                    item["content"] = self._build_responses_content(
+                        msg.get("content", "")
+                    )
+                    input_items.append(item)
 
         if instructions:
             params["instructions"] = instructions
@@ -709,6 +789,65 @@ class OpenAIClient:
 
         return params
 
+    @staticmethod
+    def _normalise_responses_image_url(image_url: Any) -> str:
+        """Return an API-fetchable URL, encoding local image files as data URLs."""
+        if isinstance(image_url, dict):
+            image_url = image_url.get("url", "")
+        if not isinstance(image_url, str) or not image_url:
+            raise ValueError("Image content must include a non-empty URL or local path")
+        if image_url.startswith(("http://", "https://", "data:")):
+            return image_url
+
+        image_path = Path(image_url).expanduser()
+        if not image_path.is_file():
+            raise FileNotFoundError(
+                f"Local image file does not exist: {image_path}"
+            )
+        mime_type = mimetypes.guess_type(image_path.name)[0]
+        if not mime_type or not mime_type.startswith("image/"):
+            raise ValueError(f"Unsupported local image type: {image_path}")
+        encoded = base64.b64encode(image_path.read_bytes()).decode("ascii")
+        return f"data:{mime_type};base64,{encoded}"
+
+    @classmethod
+    def _build_responses_content(cls, content: Any) -> Any:
+        """Translate Chat Completions content parts to Responses API parts."""
+        if not isinstance(content, list):
+            return content
+
+        responses_content: List[Any] = []
+        for part in content:
+            if not isinstance(part, dict):
+                responses_content.append(part)
+                continue
+            part_type = part.get("type")
+            if part_type == "text":
+                responses_content.append({
+                    "type": "input_text",
+                    "text": part.get("text", ""),
+                })
+            elif part_type == "image_url":
+                image_item = {
+                    "type": "input_image",
+                    "image_url": cls._normalise_responses_image_url(
+                        part.get("image_url")
+                    ),
+                }
+                image_value = part.get("image_url")
+                if isinstance(image_value, dict) and image_value.get("detail"):
+                    image_item["detail"] = image_value["detail"]
+                responses_content.append(image_item)
+            elif part_type == "input_image":
+                item = dict(part)
+                item["image_url"] = cls._normalise_responses_image_url(
+                    part.get("image_url")
+                )
+                responses_content.append(item)
+            else:
+                responses_content.append(part)
+        return responses_content
+
     def _responses_to_chat_completion(self, response) -> ChatCompletion:
         """
         Wrap a Responses API response into a ChatCompletion dataclass
@@ -890,6 +1029,8 @@ class OpenAIClient:
                         ))
                     
                     return final_response
+                except (FileNotFoundError, ValueError):
+                    raise
                 except Exception as e:
                     self.logger.warning(f"Responses API streaming failed, falling back: {e}")
                     # Fall through to Chat Completions streaming
@@ -1157,6 +1298,8 @@ class OpenAIClient:
                         ))
                     
                     return final_response
+                except (FileNotFoundError, ValueError):
+                    raise
                 except Exception as e:
                     self.logger.warning(f"Responses API async streaming failed, falling back: {e}")
                     # Fall through to Chat Completions streaming
@@ -1317,7 +1460,81 @@ class OpenAIClient:
         except Exception as e:
             self.logger.error(f"Error in async stream processing: {e}")
             return None
-    
+
+    def _track_token_usage(self, response: Any, model: str, agent_name: Optional[str] = None) -> None:
+        """Record token usage from a response into the global collector.
+
+        Usage accounting is independent from any metrics/verbose display flag:
+        a quiet default agent must still populate the public token collector
+        after a paid model call (Issue #3933). This is a cheap dict/attr read
+        plus a collector add — the expensive HTTP call is already paid.
+        """
+        try:
+            from ..telemetry.token_collector import get_token_collector, TokenMetrics
+        except Exception:
+            return
+        try:
+            usage = response.get("usage") if isinstance(response, dict) else getattr(response, "usage", None)
+            if not usage:
+                # No usage on this response (e.g. assembled stream chunks): clear
+                # any stale per-call metrics so envelopes never report a prior
+                # completion's spend as this call's usage.
+                self.last_token_metrics = None
+                return
+
+            def _val(*names: str) -> int:
+                for name in names:
+                    value = usage.get(name) if isinstance(usage, dict) else getattr(usage, name, None)
+                    if value is not None:
+                        return int(value or 0)
+                return 0
+
+            def _detail(detail_names: tuple, name: str) -> int:
+                for detail_name in detail_names:
+                    details = (
+                        usage.get(detail_name)
+                        if isinstance(usage, dict)
+                        else getattr(usage, detail_name, None)
+                    )
+                    if details:
+                        value = (
+                            details.get(name)
+                            if isinstance(details, dict)
+                            else getattr(details, name, None)
+                        )
+                        if value is not None:
+                            return int(value or 0)
+                return 0
+
+            metrics = TokenMetrics(
+                input_tokens=_val("prompt_tokens", "input_tokens"),
+                output_tokens=_val("completion_tokens", "output_tokens"),
+                cached_tokens=_val("cached_tokens") or _detail(
+                    ("prompt_tokens_details", "input_tokens_details"), "cached_tokens"
+                ),
+                reasoning_tokens=_val("reasoning_tokens") or _detail(
+                    ("completion_tokens_details", "output_tokens_details"), "reasoning_tokens"
+                ),
+                audio_input_tokens=_val("audio_input_tokens") or _detail(
+                    ("prompt_tokens_details", "input_tokens_details"), "audio_tokens"
+                ),
+                audio_output_tokens=_val("audio_output_tokens") or _detail(
+                    ("completion_tokens_details", "output_tokens_details"), "audio_tokens"
+                ),
+            )
+            if metrics.total_tokens == 0:
+                self.last_token_metrics = None
+                return
+            self.last_token_metrics = metrics
+            get_token_collector().track_tokens(
+                model=model,
+                agent=agent_name,
+                metrics=metrics,
+                metadata={"provider": "openai", "stream": False},
+            )
+        except Exception:
+            pass
+
     def create_completion(
         self,
         messages: List[Dict[str, Any]],
@@ -1352,6 +1569,8 @@ class OpenAIClient:
                 )
                 raw = self.sync_client.responses.create(**resp_params)
                 return self._responses_to_chat_completion(raw)
+            except (FileNotFoundError, ValueError):
+                raise
             except Exception as e:
                 self.logger.warning(f"Responses API failed, falling back to Chat Completions: {e}")
                 # Fall through to Chat Completions
@@ -1411,6 +1630,8 @@ class OpenAIClient:
                 )
                 raw = await self.async_client.responses.create(**resp_params)
                 return self._responses_to_chat_completion(raw)
+            except (FileNotFoundError, ValueError):
+                raise
             except Exception as e:
                 self.logger.warning(f"Responses API failed, falling back to Chat Completions: {e}")
                 # Fall through to Chat Completions
@@ -1451,6 +1672,7 @@ class OpenAIClient:
         max_iterations: int = 10,
         stream_callback: Optional[Callable] = None,
         emit_events: bool = False,
+        agent_name: Optional[str] = None,
         **kwargs
     ) -> Optional[ChatCompletion]:
         """
@@ -1487,6 +1709,11 @@ class OpenAIClient:
         # path so /stop and live steering work on the OpenAI-native loop too.
         cancel_token = kwargs.pop("cancel_token", None)
         steering_drain = kwargs.pop("steering_drain", None)
+        # Durable sink for deferred tool re-injection: the caller's persistent
+        # conversation history (e.g. the agent's ``chat_history``). Background
+        # jobs usually complete after this loop returns, so late results must
+        # land in a list a subsequent turn replays — not the per-call copy below.
+        deferred_history_sink = kwargs.pop("deferred_history_sink", None)
 
         def _is_cancelled() -> bool:
             return cancel_token is not None and getattr(cancel_token, "is_set", lambda: False)()
@@ -1616,7 +1843,13 @@ class OpenAIClient:
             
             if not final_response:
                 return None
-            
+
+            # Record usage for THIS billed completion. Every tool-loop iteration
+            # is a separate paid call, so accounting must happen per iteration —
+            # not once after the loop — or intermediate completions are dropped
+            # from session totals and by_model/by_agent rollups (Issue #3933).
+            self._track_token_usage(final_response, model, agent_name)
+
             # Trigger llm_end callback with metrics for debug output
             llm_end_time = time.perf_counter()
             llm_latency_ms = (llm_end_time - start_time) * 1000
@@ -1693,13 +1926,19 @@ class OpenAIClient:
                 # replies for this turn (keeps tool replies consecutive).
                 _deferred_media_followups = []
                 for tool_call in tool_calls:
-                    # Handle both ToolCall dataclass and OpenAI object
-                    if isinstance(tool_call, ToolCall):
-                        function_name = tool_call.function["name"]
-                        arguments = json.loads(tool_call.function["arguments"])
-                    else:
-                        function_name = tool_call.function.name
-                        arguments = json.loads(tool_call.function.arguments)
+                    # Handle both ToolCall dataclass and OpenAI object.
+                    # Guard the parse itself so malformed/truncated argument
+                    # JSON is reported back to the model instead of aborting
+                    # the whole run (matches chat_completion_with_tools_stream).
+                    function_name, arguments, _tool_call_id, err = _extract_tool_call_name_and_args(tool_call)
+                    if err is not None:
+                        logging.warning(f"Failed to parse tool call arguments: {err}")
+                        messages.append({
+                            "role": "tool",
+                            "tool_call_id": _tool_call_id,
+                            "content": json.dumps({"error": f"Invalid arguments JSON: {err}"}),
+                        })
+                        continue
                     
                     # Always trigger callback for tool call tracking (even when verbose=False)
                     display_tool_call_fn = _get_display_tool_call()
@@ -1708,7 +1947,6 @@ class OpenAIClient:
                     # Capture failures and report them back to the model instead of
                     # aborting the whole run (safe-by-default, matching the streaming
                     # path chat_completion_with_tools_stream).
-                    _tool_call_id = tool_call.id if hasattr(tool_call, 'id') else tool_call.get('id')
                     try:
                         tool_result = execute_tool_fn(
                             function_name,
@@ -1723,6 +1961,12 @@ class OpenAIClient:
                     except Exception as tool_error:
                         logging.warning(f"Tool '{function_name}' failed: {tool_error}")
                         tool_result = {"error": str(tool_error)}
+                    # Parity with the LiteLLM loop: register a deferred handle so
+                    # its eventual background value is re-injected, not lost.
+                    tool_result = _handle_native_deferred_result(
+                        tool_result, messages, _tool_call_id, function_name,
+                        history_sink=deferred_history_sink,
+                    )
                     try:
                         results_str = json.dumps(tool_result) if tool_result else "Function returned an empty output"
                     except (TypeError, ValueError):
@@ -1764,7 +2008,9 @@ class OpenAIClient:
             else:
                 # No tool calls, we're done
                 break
-        
+
+        # Usage is recorded per iteration inside the loop above so every billed
+        # completion (including intermediate tool-call rounds) is accounted for.
         return final_response
     
     async def achat_completion_with_tools(
@@ -1782,6 +2028,7 @@ class OpenAIClient:
         max_iterations: int = 10,
         stream_callback: Optional[Callable] = None,
         emit_events: bool = False,
+        agent_name: Optional[str] = None,
         **kwargs
     ) -> Optional[ChatCompletion]:
         """
@@ -1812,6 +2059,8 @@ class OpenAIClient:
         # path so /stop and live steering work on the OpenAI-native loop too.
         cancel_token = kwargs.pop("cancel_token", None)
         steering_drain = kwargs.pop("steering_drain", None)
+        # Durable sink for deferred tool re-injection (see sync counterpart).
+        deferred_history_sink = kwargs.pop("deferred_history_sink", None)
 
         def _is_cancelled() -> bool:
             return cancel_token is not None and getattr(cancel_token, "is_set", lambda: False)()
@@ -1929,6 +2178,12 @@ class OpenAIClient:
             if not final_response:
                 return None
 
+            # Record usage for THIS billed completion. Every tool-loop iteration
+            # is a separate paid call, so accounting must happen per iteration —
+            # not once after the loop — or intermediate completions are dropped
+            # from session totals and by_model/by_agent rollups (Issue #3933).
+            self._track_token_usage(final_response, model, agent_name)
+
             # Check for tool calls
             if not final_response.choices or final_response.choices[0].message is None:
                 raise ValueError("LLM returned empty or filtered response")
@@ -1979,13 +2234,19 @@ class OpenAIClient:
                 # replies for this turn (keeps tool replies consecutive).
                 _deferred_media_followups = []
                 for tool_call in tool_calls:
-                    # Handle both ToolCall dataclass and OpenAI object
-                    if isinstance(tool_call, ToolCall):
-                        function_name = tool_call.function["name"]
-                        arguments = json.loads(tool_call.function["arguments"])
-                    else:
-                        function_name = tool_call.function.name
-                        arguments = json.loads(tool_call.function.arguments)
+                    # Handle both ToolCall dataclass and OpenAI object.
+                    # Guard the parse itself so malformed/truncated argument
+                    # JSON is reported back to the model instead of aborting
+                    # the whole run (matches chat_completion_with_tools_stream).
+                    function_name, arguments, _tool_call_id, err = _extract_tool_call_name_and_args(tool_call)
+                    if err is not None:
+                        logging.warning(f"Failed to parse tool call arguments: {err}")
+                        messages.append({
+                            "role": "tool",
+                            "tool_call_id": _tool_call_id,
+                            "content": json.dumps({"error": f"Invalid arguments JSON: {err}"}),
+                        })
+                        continue
                     
                     # Always trigger callback for tool call tracking (even when verbose=False)
                     display_tool_call_fn = _get_display_tool_call()
@@ -1998,7 +2259,6 @@ class OpenAIClient:
                     # Capture failures and report them back to the model instead of
                     # aborting the whole run (safe-by-default, matching the streaming
                     # path chat_completion_with_tools_stream).
-                    _tool_call_id = tool_call.id if hasattr(tool_call, 'id') else tool_call.get('id')
                     try:
                         if asyncio.iscoroutinefunction(execute_tool_fn):
                             tool_result = await execute_tool_fn(
@@ -2032,6 +2292,12 @@ class OpenAIClient:
                         logging.warning(f"Tool '{function_name}' failed: {tool_error}")
                         tool_result = {"error": str(tool_error)}
 
+                    # Parity with the LiteLLM loop: register a deferred handle so
+                    # its eventual background value is re-injected, not lost.
+                    tool_result = _handle_native_deferred_result(
+                        tool_result, messages, _tool_call_id, function_name,
+                        history_sink=deferred_history_sink,
+                    )
                     try:
                         results_str = json.dumps(tool_result) if tool_result else "Function returned an empty output"
                     except (TypeError, ValueError):
@@ -2065,7 +2331,9 @@ class OpenAIClient:
             else:
                 # No tool calls, we're done
                 break
-        
+
+        # Usage is recorded per iteration inside the loop above so every billed
+        # completion (including intermediate tool-call rounds) is accounted for.
         return final_response
         
     def chat_completion_with_tools_stream(
@@ -2104,6 +2372,8 @@ class OpenAIClient:
         Yields:
             String chunks of the response as they are generated
         """
+        # Durable sink for deferred tool re-injection (see sync counterpart).
+        deferred_history_sink = kwargs.pop("deferred_history_sink", None)
         # Format tools for OpenAI API
         formatted_tools = self.format_tools(tools)
         
@@ -2211,16 +2481,14 @@ class OpenAIClient:
                     _deferred_media_followups = []
                     for tool_call in tool_calls:
                         # Handle both ToolCall dataclass and OpenAI object
-                        try:
-                            if isinstance(tool_call, ToolCall):
-                                function_name = tool_call.function["name"]
-                                arguments = json.loads(tool_call.function["arguments"])
-                            else:
-                                function_name = tool_call.function.name
-                                arguments = json.loads(tool_call.function.arguments)
-                        except json.JSONDecodeError as e:
+                        function_name, arguments, _, err = _extract_tool_call_name_and_args(tool_call)
+                        if err is not None:
+                            # Preserve original behaviour: only malformed JSON is
+                            # reported to the caller; other errors propagate.
+                            if not isinstance(err, json.JSONDecodeError):
+                                raise err
                             if verbose:
-                                yield f"\n[Error parsing arguments for {function_name if 'function_name' in locals() else 'unknown function'}: {str(e)}]"
+                                yield f"\n[Error parsing arguments for {function_name if function_name is not None else 'unknown function'}: {str(err)}]"
                             continue
                         
                         # Always trigger callback for tool call tracking (even when verbose=False)
@@ -2241,6 +2509,12 @@ class OpenAIClient:
                                     execute_tool_fn, iteration_count
                                 ),
                             )
+                            # Parity with the LiteLLM loop: register a deferred
+                            # handle so its eventual value is re-injected.
+                            tool_result = _handle_native_deferred_result(
+                                tool_result, messages, _tool_call_id, function_name,
+                                history_sink=deferred_history_sink,
+                            )
                             results_str = json.dumps(tool_result) if tool_result else "Function returned an empty output"
                         except ToolExecutionError:
                             raise

src/praisonai-agents/praisonaiagents/llm/unified_adapters.py

@@ -221,6 +221,7 @@ class OpenAIAdapter:
             reasoning_steps=kwargs.get('reasoning_steps', False),
             verbose=kwargs.get('verbose', True),
             max_iterations=kwargs.get('max_iterations', 10),
+            agent_name=kwargs.get('agent_name'),
             **{k: v for k, v in kwargs.items() if k not in {
                 'max_tokens', 'execute_tool_fn', 'console', 'display_fn',
                 'reasoning_steps', 'verbose', 'max_iterations',

src/praisonai-agents/praisonaiagents/lsp/__init__.py

@@ -41,6 +41,11 @@ __all__ = [
     "Range",
     # Configuration
     "LSPConfig",
+    # Auto-detection / availability helpers
+    "detect_language",
+    "detect_root_uri",
+    "probe",
+    "DEFAULT_SERVERS",
 ]
 
 
@@ -58,8 +63,9 @@ def __getattr__(name: str):
         )
         return locals()[name]
     
-    if name == "LSPConfig":
-        from .config import LSPConfig
-        return LSPConfig
+    if name in ("LSPConfig", "detect_language", "detect_root_uri", "probe",
+                "DEFAULT_SERVERS"):
+        from . import config
+        return getattr(config, name)
     
     raise AttributeError(f"module {__name__!r} has no attribute {name!r}")

src/praisonai-agents/praisonaiagents/lsp/client.py

@@ -16,7 +16,9 @@ from .types import (
     Position, Range, TextDocumentItem, TextDocumentIdentifier,
     TextDocumentPositionParams
 )
-from .config import LSPConfig
+import shutil
+
+from .config import LSPConfig, detect_root_uri, probe, path_to_uri
 
 logger = get_logger(__name__)
 
@@ -47,7 +49,8 @@ class LSPClient:
         language: str,
         command: Optional[str] = None,
         args: Optional[List[str]] = None,
-        root_uri: Optional[str] = None
+        root_uri: Optional[str] = None,
+        workspace_file: Optional[str] = None
     ):
         """
         Initialize the LSP client.
@@ -57,6 +60,10 @@ class LSPClient:
             command: Language server command (uses default if not specified)
             args: Command arguments
             root_uri: Workspace root URI
+            workspace_file: Optional file used to auto-detect the workspace root
+                from the nearest root marker (e.g. ``pyproject.toml``,
+                ``go.mod``) when ``root_uri`` is not given, so multi-root and
+                monorepo setups initialise against the correct project root.
         """
         self.config = LSPConfig(
             language=language,
@@ -64,13 +71,18 @@ class LSPClient:
             args=args or [],
             root_uri=root_uri
         )
-        
+        self._workspace_file = workspace_file
+
         self._process: Optional[asyncio.subprocess.Process] = None
         self._request_id = 0
         self._pending_requests: Dict[int, asyncio.Future] = {}
         self._diagnostics: Dict[str, List[Diagnostic]] = {}
         self._initialized = False
         self._reader_task: Optional[asyncio.Task] = None
+        # Populated with a structured, actionable message when the server
+        # binary is not on PATH, so callers can surface an explicit
+        # "unavailable + how to install" notice rather than a silent no-op.
+        self.last_error: Optional[str] = None
     
     @property
     def is_running(self) -> bool:
@@ -87,6 +99,26 @@ class LSPClient:
         if self.is_running:
             return True
         
+        # Availability pre-flight: check the *actual* configured command on
+        # PATH before spawning so a missing binary yields a single structured,
+        # actionable message instead of a raw FileNotFoundError. ``last_error``
+        # lets the caller surface an explicit "install X" notice rather than
+        # degrading silently. We gate on ``self.config.command`` (which may be a
+        # user-supplied custom server), and only surface the registry's install
+        # hint when it matches the default server for the language.
+        if not shutil.which(self.config.command):
+            _, default_command, install_hint = probe(self.config.language)
+            hint = (
+                f"; install with `{install_hint}`"
+                if install_hint and self.config.command == default_command
+                else ""
+            )
+            self.last_error = (
+                f"language server `{self.config.command}` not found on PATH{hint}"
+            )
+            logger.warning(self.last_error)
+            return False
+        
         try:
             cmd = [self.config.command] + self.config.args
             
@@ -100,8 +132,15 @@ class LSPClient:
             # Start reader task
             self._reader_task = asyncio.create_task(self._read_responses())
             
-            # Initialize the server
-            root_uri = self.config.root_uri or f"file://{os.getcwd()}"
+            # Initialize the server. Prefer an explicit root_uri, then a root
+            # detected from the workspace file's nearest root marker (so
+            # monorepos/multi-root setups initialise against the real project
+            # root), falling back to the CWD only when neither is available.
+            root_uri = self.config.root_uri
+            if root_uri is None and self._workspace_file is not None:
+                root_uri = detect_root_uri(self._workspace_file, self.config.language)
+            if root_uri is None:
+                root_uri = path_to_uri(os.getcwd())
             
             result = await self._send_request("initialize", {
                 "processId": os.getpid(),
@@ -187,7 +226,7 @@ class LSPClient:
                 logger.error(f"Failed to read file: {e}")
                 return False
         
-        uri = f"file://{os.path.abspath(file_path)}"
+        uri = path_to_uri(file_path)
         
         await self._send_notification("textDocument/didOpen", {
             "textDocument": TextDocumentItem(
@@ -205,7 +244,7 @@ class LSPClient:
         if not self._initialized:
             return
         
-        uri = f"file://{os.path.abspath(file_path)}"
+        uri = path_to_uri(file_path)
         
         await self._send_notification("textDocument/didClose", {
             "textDocument": TextDocumentIdentifier(uri=uri).to_dict()
@@ -221,7 +260,7 @@ class LSPClient:
         Returns:
             List of diagnostics
         """
-        uri = f"file://{os.path.abspath(file_path)}"
+        uri = path_to_uri(file_path)
         return self._diagnostics.get(uri, [])
     
     async def get_completions(
@@ -244,7 +283,7 @@ class LSPClient:
         if not self._initialized:
             return []
         
-        uri = f"file://{os.path.abspath(file_path)}"
+        uri = path_to_uri(file_path)
         
         result = await self._send_request("textDocument/completion", {
             "textDocument": {"uri": uri},
@@ -281,7 +320,7 @@ class LSPClient:
         if not self._initialized:
             return []
         
-        uri = f"file://{os.path.abspath(file_path)}"
+        uri = path_to_uri(file_path)
         
         result = await self._send_request("textDocument/definition", {
             "textDocument": {"uri": uri},
@@ -320,7 +359,7 @@ class LSPClient:
         if not self._initialized:
             return []
         
-        uri = f"file://{os.path.abspath(file_path)}"
+        uri = path_to_uri(file_path)
         
         result = await self._send_request("textDocument/references", {
             "textDocument": {"uri": uri},
@@ -353,7 +392,7 @@ class LSPClient:
         if not self._initialized:
             return None
         
-        uri = f"file://{os.path.abspath(file_path)}"
+        uri = path_to_uri(file_path)
         
         result = await self._send_request("textDocument/hover", {
             "textDocument": {"uri": uri},
@@ -390,7 +429,7 @@ class LSPClient:
         if not self._initialized:
             return []
         
-        uri = f"file://{os.path.abspath(file_path)}"
+        uri = path_to_uri(file_path)
         
         result = await self._send_request("textDocument/documentSymbol", {
             "textDocument": {"uri": uri}

src/praisonai-agents/praisonaiagents/lsp/config.py

@@ -2,35 +2,124 @@
 LSP Configuration for PraisonAI Agents.
 """
 
+import os
+import shutil
 from dataclasses import dataclass, field
-from typing import Optional, Dict, List
+from pathlib import Path
+from typing import Optional, Dict, List, Tuple
 
 
-# Default language server commands
+# Data-driven language server registry.  Each entry carries the spawn command
+# plus enough metadata for zero-config auto-detection:
+#   - command / args: how to start the server
+#   - extensions: file extensions that map to this language
+#   - root_markers: files/dirs that mark a workspace root (nearest wins), so
+#     ``rootUri`` is the real project root rather than ``os.getcwd()``
+#   - install_hint: an actionable "how to install" string surfaced when the
+#     server binary is absent, so degradation is explicit instead of silent
 DEFAULT_SERVERS = {
     "python": {
         "command": "pylsp",
         "args": [],
+        "extensions": [".py", ".pyi"],
+        "root_markers": ["pyproject.toml", "setup.py", "setup.cfg", "requirements.txt", ".git"],
+        "install_hint": "pip install python-lsp-server",
     },
     "javascript": {
         "command": "typescript-language-server",
         "args": ["--stdio"],
+        "extensions": [".js", ".jsx", ".mjs", ".cjs"],
+        "root_markers": ["package.json", "tsconfig.json", "jsconfig.json", ".git"],
+        "install_hint": "npm install -g typescript-language-server typescript",
     },
     "typescript": {
         "command": "typescript-language-server",
         "args": ["--stdio"],
+        "extensions": [".ts", ".tsx"],
+        "root_markers": ["tsconfig.json", "package.json", "jsconfig.json", ".git"],
+        "install_hint": "npm install -g typescript-language-server typescript",
     },
     "rust": {
         "command": "rust-analyzer",
         "args": [],
+        "extensions": [".rs"],
+        "root_markers": ["Cargo.toml", "Cargo.lock", ".git"],
+        "install_hint": "rustup component add rust-analyzer",
     },
     "go": {
         "command": "gopls",
         "args": [],
+        "extensions": [".go"],
+        "root_markers": ["go.mod", "go.sum", ".git"],
+        "install_hint": "go install golang.org/x/tools/gopls@latest",
     },
 }
 
 
+def _extension_map() -> Dict[str, str]:
+    """Return a ``{extension: language}`` map derived from the registry."""
+    mapping: Dict[str, str] = {}
+    for language, server in DEFAULT_SERVERS.items():
+        for ext in server.get("extensions", []):
+            mapping.setdefault(ext.lower(), language)
+    return mapping
+
+
+def detect_language(file_path: str) -> Optional[str]:
+    """Resolve the LSP language id for ``file_path`` by extension, else ``None``."""
+    return _extension_map().get(os.path.splitext(file_path)[1].lower())
+
+
+def path_to_uri(path: str) -> str:
+    """Convert an absolute filesystem *path* to a valid ``file://`` URI.
+
+    Uses :meth:`pathlib.Path.as_uri`, which percent-encodes reserved characters
+    (spaces, ``#`` …) so paths with such characters produce a valid, non-truncated
+    URI that language servers accept.
+    """
+    return Path(os.path.abspath(path)).as_uri()
+
+
+def detect_root_uri(file_path: str, language: Optional[str] = None) -> Optional[str]:
+    """Discover the workspace root for ``file_path`` from the nearest root marker.
+
+    Walks up from the file's directory looking for a language-appropriate root
+    marker (``pyproject.toml``, ``go.mod``, ``Cargo.toml``, ``tsconfig.json``,
+    ``.git`` …).  Returns a ``file://`` URI for the nearest matching directory,
+    or ``None`` when no marker is found so the caller can fall back to the CWD.
+    """
+    language = language or detect_language(file_path)
+    server = DEFAULT_SERVERS.get(language) if language else None
+    markers = server.get("root_markers", []) if server else [".git"]
+    start = os.path.dirname(os.path.abspath(file_path))
+    current = start
+    while True:
+        for marker in markers:
+            if os.path.exists(os.path.join(current, marker)):
+                return path_to_uri(current)
+        parent = os.path.dirname(current)
+        if parent == current:
+            return None
+        current = parent
+
+
+def probe(language: str) -> Tuple[bool, Optional[str], Optional[str]]:
+    """Check whether the language server for ``language`` is on ``PATH``.
+
+    Returns ``(available, command, install_hint)``.  ``available`` is ``True``
+    when the configured server binary is found via ``shutil.which``; otherwise
+    ``command`` and ``install_hint`` describe what is missing and how to install
+    it so the degradation can be surfaced instead of silently swallowed.
+    """
+    server = DEFAULT_SERVERS.get(language)
+    if not server:
+        return False, None, None
+    command = server["command"]
+    if shutil.which(command):
+        return True, command, server.get("install_hint")
+    return False, command, server.get("install_hint")
+
+
 @dataclass
 class LSPConfig:
     """Configuration for LSP client."""
@@ -47,6 +136,6 @@ class LSPConfig:
             if self.language in DEFAULT_SERVERS:
                 server = DEFAULT_SERVERS[self.language]
                 self.command = server["command"]
-                self.args = server["args"]
+                self.args = list(server["args"])
             else:
                 raise ValueError(f"No default server for language: {self.language}")

src/praisonai-agents/praisonaiagents/main.py

@@ -742,36 +742,8 @@ class ReflectionOutput(BaseModel):
     reflection: str
     satisfactory: Literal["yes", "no"]
 
-class TaskOutput(BaseModel):
-    model_config = ConfigDict(arbitrary_types_allowed=True)
-    description: str
-    summary: Optional[str] = None
-    raw: str
-    pydantic: Optional[BaseModel] = None
-    json_dict: Optional[Dict[str, Any]] = None
-    agent: str
-    output_format: Literal["RAW", "JSON", "Pydantic"] = "RAW"
-    token_metrics: Optional['TokenMetrics'] = None  # Add token metrics field
-    callback_error: Optional[str] = None
-    non_fatal_errors: Optional[list[str]] = None
-
-    def json(self) -> Optional[str]:
-        if self.output_format == "JSON" and self.json_dict:
-            return json.dumps(self.json_dict)
-        return None
-
-    def to_dict(self) -> dict:
-        output_dict = {}
-        if self.json_dict:
-            output_dict.update(self.json_dict)
-        if self.pydantic:
-            output_dict.update(self.pydantic.model_dump())
-        return output_dict
-
-    def __str__(self):
-        if self.pydantic:
-            return str(self.pydantic)
-        elif self.json_dict:
-            return json.dumps(self.json_dict)
-        else:
-            return self.raw 
+# Re-export the single canonical TaskOutput so main.TaskOutput and
+# output.models.TaskOutput are the SAME class. Previously two near-identical
+# classes both named TaskOutput existed, which broke guardrail validation
+# (GuardrailResult expected one class while agents produced the other).
+from .output.models import TaskOutput

src/praisonai-agents/praisonaiagents/managed/_compute_bridge.py

@@ -11,30 +11,83 @@ from __future__ import annotations
 import importlib
 from typing import Any, Optional
 
-_INSTALL_HINT = "Remote compute requires the wrapper package: pip install praisonai"
+_INSTALL_HINT = "Remote compute requires: pip install praisonai-sandbox"
 
 # name -> (module, attribute). Kept in sync with
 # praisonai.integrations.managed_local._resolve_compute
 _PROVIDERS = {
-    "local": ("praisonai.integrations.compute.local", "LocalCompute"),
-    "docker": ("praisonai.integrations.compute.docker", "DockerCompute"),
-    "e2b": ("praisonai.integrations.compute.e2b", "E2BCompute"),
-    "modal": ("praisonai.integrations.compute.modal_compute", "ModalCompute"),
-    "daytona": ("praisonai.integrations.compute.daytona", "DaytonaCompute"),
-    "flyio": ("praisonai.integrations.compute.flyio", "FlyioCompute"),
-    "tenki": ("praisonai.integrations.compute.tenki", "TenkiCompute"),
+    "local": ("praisonai_sandbox.compute.local", "LocalCompute"),
+    "docker": ("praisonai_sandbox.compute.docker", "DockerCompute"),
+    "e2b": ("praisonai_sandbox.compute.e2b", "E2BCompute"),
+    "modal": ("praisonai_sandbox.compute.modal_compute", "ModalCompute"),
+    "daytona": ("praisonai_sandbox.compute.daytona", "DaytonaCompute"),
+    "flyio": ("praisonai_sandbox.compute.flyio", "FlyioCompute"),
+    "tenki": ("praisonai_sandbox.compute.tenki", "TenkiCompute"),
 }
 
+# What to actually type when a provider is missing. Every name here must point
+# at an extra that EXISTS and installs something -- three of these used to name
+# extras that were absent or empty, so a user followed the official hint, waited
+# for the install, and hit the identical failure again.
+#
+# The providers in _PROVIDERS now live in praisonai-sandbox, so the failing
+# import is `praisonai_sandbox.compute.*`: point the hint at the lightweight
+# package that owns the module, not the heavy wrapper. Reaching E2B through
+# praisonai-sandbox[e2b] resolves 58 packages against 143 through praisonai.
+# Sandbox-only places (novita/sandlock/ssh) resolve through the
+# SandboxComputeAdapter and only need `praisonai-sandbox`.
 _EXTRA_HINTS = {
-    "e2b": "pip install praisonai[e2b]",
-    "modal": "pip install praisonai[modal]",
-    "daytona": "pip install praisonai[daytona]",
+    "e2b": "pip install 'praisonai-sandbox[e2b]'",
+    "modal": "pip install 'praisonai-sandbox[modal]'",
+    "daytona": "pip install 'praisonai-sandbox[daytona]'",
+    "docker": "pip install 'praisonai-sandbox[docker]'",
+    "novita": "pip install 'praisonai-sandbox[novita]'",
+    "sandlock": "pip install 'praisonai-sandbox[sandlock]'",
+    "ssh": "pip install 'praisonai-sandbox[ssh]'",
+    # flyio and tenki need a credential rather than an SDK; flyio speaks plain
+    # HTTP through aiohttp, which praisonaiagents already requires. Export the
+    # variable so a child process inherits it -- a bare assignment would not.
+    "flyio": "export FLY_API_TOKEN=... (see https://fly.io/docs/flyctl/auth-token)",
+    "tenki": "pip install 'praisonai-sandbox[tenki]'",
 }
 
 
+# Places that live only in praisonai-sandbox. They speak SandboxProtocol
+# (one object IS one environment) rather than ComputeProviderProtocol (one
+# object hands out many), so an adapter supplies the missing half.
+from ._sandbox_adapter import NEEDS_INSTANCE, SANDBOX_ONLY  # noqa: E402
+
+# Spellings users already have in their code, mapped to the canonical name.
+_ALIASES = {"native": "sandlock"}
+
+
+def _entry_point_providers() -> dict:
+    """Places contributed by other installed packages.
+
+    The sandbox and managed-backend registries already discover plugins this
+    way (``praisonai.sandbox``, ``praisonai.managed_backends``); the compute
+    registry was the one hardcoded dict, so adding a place meant editing core.
+    A provider can now ship in its own distribution:
+
+        [project.entry-points."praisonai.compute"]
+        runpod = "praisonai_compute_runpod:RunpodCompute"
+    """
+    found = {}
+    try:
+        from importlib.metadata import entry_points
+
+        for ep in entry_points(group="praisonai.compute"):
+            found[ep.name.lower()] = ep
+    except Exception:  # pragma: no cover - discovery must never break resolution
+        pass
+    return found
+
+
 def available_providers() -> list:
-    """Return the compute provider names this bridge can resolve."""
-    return sorted(_PROVIDERS)
+    """Every place ``tools_run_on=`` / ``run_in=`` accepts, from all registries."""
+    return sorted(
+        set(_PROVIDERS) | set(SANDBOX_ONLY) | set(_entry_point_providers())
+    )
 
 
 def compute_package_available() -> bool:
@@ -58,15 +111,44 @@ def resolve_compute(compute: Optional[Any]) -> Optional[Any]:
     if not isinstance(compute, str):
         if hasattr(compute, "provision") and hasattr(compute, "execute"):
             return compute
+        # A SandboxProtocol instance -- e.g. SSHSandbox(host=...) -- is adapted,
+        # which is the only way to pass connection details.
+        if hasattr(compute, "run_command") and hasattr(compute, "start"):
+            from ._sandbox_adapter import wrap_sandbox_instance
+
+            return wrap_sandbox_instance(compute)
         raise TypeError(
             f"compute= expects a provider name or a ComputeProviderProtocol "
             f"instance, got {type(compute).__name__}"
         )
 
     name = compute.lower().strip()
+    name = _ALIASES.get(name, name)
+
+    # A sandbox-only place: adapt it to the compute-provider shape.
+    if name in SANDBOX_ONLY:
+        if name in NEEDS_INSTANCE:
+            raise ValueError(
+                f"{name!r} needs connection details a bare name cannot carry. "
+                f"Instead, {NEEDS_INSTANCE[name]}"
+            )
+        from ._sandbox_adapter import SandboxComputeAdapter
+
+        return SandboxComputeAdapter(name)
+
+    external = _entry_point_providers()
+    if name in external and name not in _PROVIDERS:
+        # Let a contributed place describe itself. Without this the repr falls
+        # back to the bare name, because the phrase table is a literal that
+        # cannot know about packages installed later.
+        _remember_display_name(name, external[name])
+        # A place contributed by another package. Loaded on demand, so an
+        # installed-but-broken plugin cannot break anything that never names it.
+        return external[name].load()()
+
     if name not in _PROVIDERS:
         raise ValueError(
-            f"Unknown compute provider {name!r}. "
+            f"Unknown place {name!r}. "
             f"Available: {', '.join(available_providers())}"
         )
 
@@ -78,3 +160,23 @@ def resolve_compute(compute: Optional[Any]) -> Optional[Any]:
         raise ImportError(f"compute={name!r} requires {module_name}. {hint}") from exc
 
     return getattr(module, attr)()
+
+
+#: Display phrases contributed by external providers, filled in on resolve.
+_DISPLAY_NAMES: dict = {}
+
+
+def _remember_display_name(name: str, entry_point) -> None:
+    """Record a contributed provider's own phrase, if it declares one."""
+    try:
+        cls = entry_point.load()
+        phrase = getattr(cls, "display_name", None)
+        if isinstance(phrase, str) and phrase:
+            _DISPLAY_NAMES[name] = phrase
+    except Exception:  # pragma: no cover - naming must never break resolution
+        pass
+
+
+def contributed_display_name(name: str):
+    """The phrase a contributed place declared for itself, if any."""
+    return _DISPLAY_NAMES.get((name or "").lower())

src/praisonai-agents/praisonaiagents/managed/_sandbox_adapter.py

@@ -0,0 +1,254 @@
+"""Expose a single sandbox environment as a one-instance compute provider.
+
+There are two execution stacks in this project and they are *not* duplicates:
+
+* ``SandboxProtocol`` (praisonai-sandbox) -- one object **is** one environment:
+  ``start / stop / execute / run_command / reset``.
+* ``ComputeProviderProtocol`` -- one object **hands out** many environments:
+  ``provision / execute(instance_id, ...) / shutdown / list_instances``.
+
+``tools_run_on=`` speaks the second protocol, so four backends that only exist
+as sandboxes -- ``subprocess``, ``sandlock``, ``ssh``, ``novita`` -- were
+unreachable through it. This adapter is the missing half: it holds exactly one
+environment and presents the manager-shaped API over it.
+
+Deliberate limitation: ``list_instances()`` can only report what *this process*
+provisioned, so adapter-backed places never appear in ``praisonai managed ps``.
+A single environment has no cross-process registry to consult, and pretending
+otherwise would be worse than saying so.
+"""
+
+from __future__ import annotations
+
+import logging
+import uuid
+from typing import Any, Dict, List, Optional
+
+logger = logging.getLogger(__name__)
+
+# Backends that exist only as SandboxProtocol implementations. The literal is
+# the floor, not the ceiling: the sandbox registry accepts plugins by entry
+# point, so a hardcoded tuple silently withheld contributed backends from
+# tools_run_on=. `capsule` was the live example -- it ships from
+# praisonai-plugins, worked with run_in=, and was invisible to tools_run_on=.
+_SANDBOX_ONLY_FLOOR = ("subprocess", "sandlock", "ssh", "novita")
+
+
+def sandbox_only_names() -> tuple:
+    """Sandbox backends with no compute-registry equivalent.
+
+    Anything the compute registry already provides is excluded, so a name that
+    exists in both (docker, e2b, modal, daytona) keeps its richer compute
+    implementation rather than being downgraded to the one-instance adapter.
+    """
+    names = set(_SANDBOX_ONLY_FLOOR)
+    try:
+        from ..sandbox._sandbox_bridge import get_sandbox_registry
+
+        registry = get_sandbox_registry()
+        registry = registry.default() if isinstance(registry, type) else registry
+        names |= {str(n).lower() for n in registry.list_all_names()}
+    except Exception:  # pragma: no cover - discovery must never break resolution
+        pass
+
+    try:
+        from ._compute_bridge import _PROVIDERS
+
+        names -= set(_PROVIDERS)
+    except Exception:
+        pass
+    return tuple(sorted(names))
+
+
+class _SandboxOnly(tuple):
+    """A tuple that re-reads the registry, so `in` sees plugins installed later.
+
+    Kept as a tuple subclass because SANDBOX_ONLY is imported and iterated in
+    several places that predate this.
+    """
+
+    def __contains__(self, item):
+        return item in sandbox_only_names()
+
+    def __iter__(self):
+        return iter(sandbox_only_names())
+
+    def __len__(self):
+        return len(sandbox_only_names())
+
+
+SANDBOX_ONLY = _SandboxOnly(_SANDBOX_ONLY_FLOOR)
+
+# Names that need connection details no bare string can carry.
+NEEDS_INSTANCE = {
+    "ssh": (
+        "run on a remote machine over SSH by passing the object, so it can "
+        "carry the host: SSHSandbox(host='...', user='...')"
+    ),
+}
+
+
+class SandboxComputeAdapter:
+    """One ``SandboxProtocol`` environment, wearing the compute-provider shape."""
+
+    def __init__(self, sandbox_type: str, config: Optional[Any] = None,
+                 sandbox: Optional[Any] = None):
+        self._type = sandbox_type
+        self._config = config
+        self._sandbox = sandbox          # pre-built instance, if supplied
+        self._owns_sandbox = sandbox is None
+        self._instance_id: Optional[str] = None
+
+    # ── identity ─────────────────────────────────────────────────────────────
+    @property
+    def provider_name(self) -> str:
+        return self._type
+
+    @property
+    def is_available(self) -> bool:
+        if self._sandbox is not None:
+            return bool(getattr(self._sandbox, "is_available", True))
+        try:
+            from ..sandbox._sandbox_bridge import resolve_sandbox_class
+
+            cls = resolve_sandbox_class(self._type)
+            probe = cls() if self._type not in NEEDS_INSTANCE else None
+            return bool(getattr(probe, "is_available", False)) if probe else False
+        except Exception:
+            return False
+
+    # ── lifecycle ────────────────────────────────────────────────────────────
+    async def provision(self, config: Any) -> Any:
+        from .protocols import InstanceInfo, InstanceStatus
+
+        if self._sandbox is None:
+            from ..sandbox import SandboxConfig
+            from ..sandbox.manager import SandboxManager
+
+            cfg = self._config
+            if cfg is None:
+                # "/workspace" is the convention for *container* sandboxes and
+                # does not exist on a Mac or a bare Linux box -- creating it
+                # fails with "Read-only file system". Local backends pick their
+                # own private temp dir, so only pass a working_dir through when
+                # the caller actually chose one.
+                wanted_dir = getattr(config, "working_dir", None)
+                local_place = self._type in ("subprocess", "sandlock")
+                if local_place and wanted_dir in (None, "", "/workspace"):
+                    wanted_dir = None
+                cfg = SandboxConfig(
+                    sandbox_type=self._type,
+                    env=dict(getattr(config, "env", None) or {}),
+                    **({"working_dir": wanted_dir} if wanted_dir else {}),
+                )
+                # This sandbox exists to back the model's execute_command tool,
+                # which IS a shell. The default policy sets allow_subprocess=
+                # False, which makes `sh -c` -- and therefore every pipe,
+                # redirect and && the model writes -- fail as "disabled by
+                # security policy". Blocking the shell inside a shell tool is a
+                # contradiction, not a protection; blocked_commands and
+                # blocked_paths still apply.
+                if getattr(cfg, "security_policy", None) is not None:
+                    cfg.security_policy.allow_subprocess = True
+            if getattr(config, "image", None):
+                cfg.image = config.image
+            self._manager = SandboxManager(cfg)
+            self._sandbox = await self._manager.__aenter__()
+        else:
+            self._manager = None
+            starter = getattr(self._sandbox, "start", None)
+            if starter is not None:
+                await starter()
+
+        self._instance_id = f"{self._type}_{uuid.uuid4().hex[:12]}"
+        logger.info("[sandbox_adapter] provisioned %s as %s", self._type, self._instance_id)
+        return InstanceInfo(
+            instance_id=self._instance_id,
+            status=InstanceStatus.RUNNING,
+            endpoint=f"{self._type}://local",
+            provider=self._type,
+            metadata={"adapter": True, "discoverable": False},
+        )
+
+    async def execute(self, instance_id: str, command: str,
+                      timeout: Optional[int] = 300) -> Dict[str, Any]:
+        if self._sandbox is None:
+            raise RuntimeError(
+                f"{self._type} sandbox is not running; provision() was not called"
+            )
+        # The bridged execute_command tool hands us a shell command string, and
+        # models write pipes, redirects and && constantly. Without shell=True
+        # `echo x > f` is argv-split and "> f" arrives as a literal argument.
+        try:
+            result = await self._sandbox.run_command(command, shell=True)
+        except TypeError:
+            # Backends predating the shell= parameter.
+            result = await self._sandbox.run_command(command)
+        # A policy denial arrives as result.error with exit_code=None. Dropping
+        # it would turn "blocked by security policy" into a silent success --
+        # the caller would see exit 0 and empty output.
+        error = getattr(result, "error", None)
+        exit_code = getattr(result, "exit_code", None)
+        stderr = getattr(result, "stderr", "") or ""
+        if error:
+            stderr = f"{stderr}\n{error}".strip() if stderr else str(error)
+            if exit_code in (None, 0):
+                exit_code = 1
+        return {
+            "stdout": getattr(result, "stdout", "") or "",
+            "stderr": stderr,
+            "exit_code": exit_code or 0,
+        }
+
+    async def shutdown(self, instance_id: str) -> None:
+        try:
+            if getattr(self, "_manager", None) is not None:
+                await self._manager.__aexit__(None, None, None)
+            elif self._sandbox is not None:
+                stopper = getattr(self._sandbox, "stop", None)
+                if stopper is not None:
+                    await stopper()
+        finally:
+            if self._owns_sandbox:
+                self._sandbox = None
+            self._manager = None
+            self._instance_id = None
+
+    async def get_status(self, instance_id: str) -> Any:
+        from .protocols import InstanceInfo, InstanceStatus
+
+        running = self._sandbox is not None and instance_id == self._instance_id
+        return InstanceInfo(
+            instance_id=instance_id,
+            status=InstanceStatus.RUNNING if running else InstanceStatus.STOPPED,
+            provider=self._type,
+        )
+
+    async def list_instances(self) -> List[Any]:
+        """Only what this process provisioned -- see the module docstring."""
+        from .protocols import InstanceInfo, InstanceStatus
+
+        if self._instance_id is None:
+            return []
+        return [InstanceInfo(
+            instance_id=self._instance_id,
+            status=InstanceStatus.RUNNING,
+            provider=self._type,
+            metadata={"adapter": True, "discoverable": False},
+        )]
+
+    # ── file transfer ────────────────────────────────────────────────────────
+    async def upload_file(self, instance_id: str, local_path: str, remote_path: str) -> None:
+        with open(local_path, "r") as handle:
+            await self._sandbox.write_file(remote_path, handle.read())
+
+    async def download_file(self, instance_id: str, remote_path: str, local_path: str) -> None:
+        content = await self._sandbox.read_file(remote_path)
+        with open(local_path, "w") as handle:
+            handle.write(content if isinstance(content, str) else str(content))
+
+
+def wrap_sandbox_instance(sandbox: Any) -> SandboxComputeAdapter:
+    """Adapt an already-constructed sandbox (e.g. ``SSHSandbox(host=...)``)."""
+    name = getattr(sandbox, "sandbox_type", None) or type(sandbox).__name__.lower()
+    return SandboxComputeAdapter(str(name), sandbox=sandbox)

src/praisonai-agents/praisonaiagents/managed/shared_compute.py

@@ -17,6 +17,7 @@ import logging
 import shlex
 import threading
 import uuid
+import weakref
 from typing import Any, Dict, List, Optional
 
 from ._compute_bridge import resolve_compute
@@ -118,8 +119,13 @@ class SharedCompute:
 
     def shutdown(self) -> None:
         """Tear down the instance and restore any agents we patched."""
-        for agent, original_tools, original_backstory in self._patched:
+        for ref, original_tools, original_backstory in self._patched:
+            agent = ref()
+            if agent is None:
+                continue          # already collected; nothing left to restore
             try:
+                if getattr(agent, "_tools_sandbox", None) is self:
+                    agent._tools_sandbox = None
                 agent.tools = original_tools
                 agent.backstory = original_backstory
                 cache = getattr(agent, "_system_prompt_cache", None)
@@ -136,6 +142,8 @@ class SharedCompute:
             logger.info("[shared_compute] shut down instance %s", self.instance_id)
         except Exception as exc:  # pragma: no cover - teardown must not mask errors
             logger.warning("[shared_compute] shutdown failed for %s: %s", self.instance_id, exc)
+            self.instance_id = None
+            raise            # the caller logs whether the sandbox was released
         finally:
             self.instance_id = None
 
@@ -187,9 +195,14 @@ class SharedCompute:
             # heredoc early (which would truncate the file and run the remainder
             # as shell commands). Guard against the astronomically-unlikely case
             # where the token still appears in the body.
-            delimiter = f"__PRAISON_EOF_{uuid.uuid4().hex}__"
+            # Uppercase hex on purpose. uuid4().hex is lowercase, and the
+            # security policy substring-matches blocked_commands against the
+            # whole command -- "dd" is one of them, and appears in 9.6% of
+            # random lowercase hex strings. One write in ten failed with
+            # "Blocked command pattern: dd" for no reason the caller could see.
+            delimiter = f"__PRAISON_EOF_{uuid.uuid4().hex.upper()}__"
             while delimiter in content:
-                delimiter = f"__PRAISON_EOF_{uuid.uuid4().hex}__"
+                delimiter = f"__PRAISON_EOF_{uuid.uuid4().hex.upper()}__"
             heredoc = (
                 f"mkdir -p $(dirname {quoted}) && "
                 f"cat > {quoted} <<'{delimiter}'\n{content}\n{delimiter}"
@@ -203,13 +216,25 @@ class SharedCompute:
             """List files in a directory on the shared remote sandbox."""
             return self._format(self.run_command(f"ls -la {shlex.quote(path)}"))
 
-        return [execute_command, read_file, write_file, list_files]
+        bridged = [execute_command, read_file, write_file, list_files]
+        # Tag them so Agent.clone_for_channel() drops them: each closes over
+        # THIS sandbox, so a clone inheriting them would talk to the original
+        # agent's instance and then attach a second sandbox of its own --
+        # duplicating the capability prompt and splitting its work across two
+        # machines. The clone regenerates its own on first use.
+        for tool in bridged:
+            tool._praison_sandbox_tool = True
+        return bridged
 
-    def attach(self, agents: List[Any]) -> None:
+    def attach(self, agents: List[Any], *, override_declared: bool = False) -> None:
         """Give each agent the shared tools and tell it they exist.
 
         Agents that already carry their own managed ``backend`` are skipped --
         they have deliberately been pointed at their own runtime.
+
+        ``override_declared`` is set only by the agent's *own* placement, which
+        is this sandbox's whole reason for existing. A run-level sandbox leaves
+        it False so an agent that named its own place keeps it.
         """
         tools = self.build_tools()
         by_name = {t.__name__: t for t in tools}
@@ -217,6 +242,31 @@ class SharedCompute:
         for agent in agents:
             if agent is None or getattr(agent, "backend", None) is not None:
                 continue
+            # An agent that named its own place, or already has a sandbox, must
+            # not be attached again. A second attach appends CAPABILITY_PROMPT
+            # twice and, if the two teardowns do not happen in LIFO order,
+            # permanently restores the wrong tools: the agent is left holding
+            # four tools bound to a shut-down instance.
+            #
+            # `tools_run_on` is checked as well as `_tools_sandbox` because of
+            # ordering: a flow attaches at the start of the run, before the
+            # agent's first chat() has created its own sandbox. Checking only
+            # the live sandbox would attach here and let the agent attach again
+            # on its first turn. An explicit per-agent choice outranks the run's
+            # default, so the flow leaves it alone.
+            declared_own = (
+                not override_declared
+                and getattr(agent, "tools_run_on", None) is not None
+            )
+            if getattr(agent, "_tools_sandbox", None) is not None or declared_own:
+                logger.debug(
+                    "[shared_compute] %r already runs its tools on %s; leaving it alone",
+                    getattr(agent, "name", agent),
+                    getattr(agent, "tools_run_on", "its own sandbox"),
+                )
+                continue
+            if any(agent is ref() for ref, _, _ in self._patched):
+                continue
 
             original_tools = list(getattr(agent, "tools", None) or [])
             # Agent.__init__ copies `instructions` into role/goal/backstory and
@@ -248,4 +298,18 @@ class SharedCompute:
                 logger.warning("[shared_compute] could not attach tools to %r: %s", agent, exc)
                 continue
 
-            self._patched.append((agent, original_tools, original_backstory))
+            # Mark where this agent's tools now run, so a nested attach (or a
+            # later ensure_tools_placed) can see it is already bound.
+            try:
+                agent._tools_sandbox = self
+            except Exception:  # pragma: no cover - exotic agent objects
+                pass
+            # Weakly: an agent points at its SharedCompute and the SharedCompute
+            # pointed back, so a dropped agent was reclaimed as a cycle and its
+            # sandbox was never shut down. Holding the agent weakly lets
+            # weakref.finalize see the agent die and release the sandbox.
+            try:
+                ref = weakref.ref(agent)
+            except TypeError:  # pragma: no cover - non-weakrefable agent
+                ref = lambda _agent=agent: _agent
+            self._patched.append((ref, original_tools, original_backstory))

src/praisonai-agents/praisonaiagents/mcp/__init__.py

@@ -11,94 +11,62 @@ using different transport methods:
 Protocol Revision: 2025-11-25
 """
 from .mcp import MCP
+from .._lazy import create_lazy_getattr
+
+
+def _t(module_path: str, attr_name: str):
+    """Build a (module_path, attr_name) lazy-import target.
+
+    Using a helper instead of bare 2-string tuple literals keeps secret
+    scanners from misclassifying the auth/oauth import targets below as
+    hardcoded "Authentication Tuple" credentials (they are import paths,
+    not secrets).
+    """
+    return (module_path, attr_name)
+
 
 # Lazy imports for optional components
-def __getattr__(name):
-    """Lazy import for optional modules."""
-    if name == "WebSocketMCPClient":
-        from .mcp_websocket import WebSocketMCPClient
-        return WebSocketMCPClient
-    elif name == "SessionManager":
-        from .mcp_session import SessionManager
-        return SessionManager
-    elif name == "SecurityConfig":
-        from .mcp_security import SecurityConfig
-        return SecurityConfig
-    elif name == "BaseTransport":
-        from .mcp_transport import BaseTransport
-        return BaseTransport
-    elif name == "TransportConfig":
-        from .mcp_transport import TransportConfig
-        return TransportConfig
-    elif name == "ToolsMCPServer":
-        from .mcp_server import ToolsMCPServer
-        return ToolsMCPServer
-    elif name == "launch_tools_mcp_server":
-        from .mcp_server import launch_tools_mcp_server
-        return launch_tools_mcp_server
-    elif name == "function_to_mcp_schema":
-        from .mcp_utils import function_to_mcp_schema
-        return function_to_mcp_schema
-    elif name == "get_tool_metadata":
-        from .mcp_utils import get_tool_metadata
-        return get_tool_metadata
-    elif name == "python_type_to_json_schema":
-        from .mcp_utils import python_type_to_json_schema
-        return python_type_to_json_schema
-    elif name == "fix_array_schemas":
-        from .mcp_schema_utils import fix_array_schemas
-        return fix_array_schemas
-    elif name == "ThreadLocalEventLoop":
-        from .mcp_schema_utils import ThreadLocalEventLoop
-        return ThreadLocalEventLoop
-    elif name == "get_thread_local_event_loop":
-        from .mcp_schema_utils import get_thread_local_event_loop
-        return get_thread_local_event_loop
-    # Auth storage (lazy loaded)
-    elif name == "MCPAuthStorage":
-        from .mcp_auth_storage import MCPAuthStorage
-        return MCPAuthStorage
-    elif name == "get_default_auth_filepath":
-        from .mcp_auth_storage import get_default_auth_filepath
-        return get_default_auth_filepath
-    # OAuth callback utilities (lazy loaded)
-    elif name == "OAuthCallbackHandler":
-        from .mcp_oauth_callback import OAuthCallbackHandler
-        return OAuthCallbackHandler
-    elif name == "generate_state":
-        from .mcp_oauth_callback import generate_state
-        return generate_state
-    elif name == "generate_code_verifier":
-        from .mcp_oauth_callback import generate_code_verifier
-        return generate_code_verifier
-    elif name == "generate_code_challenge":
-        from .mcp_oauth_callback import generate_code_challenge
-        return generate_code_challenge
-    elif name == "get_redirect_url":
-        from .mcp_oauth_callback import get_redirect_url
-        return get_redirect_url
-    elif name == "OAUTH_CALLBACK_PORT":
-        from .mcp_oauth_callback import OAUTH_CALLBACK_PORT
-        return OAUTH_CALLBACK_PORT
-    elif name == "OAUTH_CALLBACK_PATH":
-        from .mcp_oauth_callback import OAUTH_CALLBACK_PATH
-        return OAUTH_CALLBACK_PATH
+_LAZY_IMPORTS = {
+    "WebSocketMCPClient": _t("praisonaiagents.mcp.mcp_websocket", "WebSocketMCPClient"),
+    "SessionManager": _t("praisonaiagents.mcp.mcp_session", "SessionManager"),
+    "SecurityConfig": _t("praisonaiagents.mcp.mcp_security", "SecurityConfig"),
+    "BaseTransport": _t("praisonaiagents.mcp.mcp_transport", "BaseTransport"),
+    "TransportConfig": _t("praisonaiagents.mcp.mcp_transport", "TransportConfig"),
+    "ToolsMCPServer": _t("praisonaiagents.mcp.mcp_server", "ToolsMCPServer"),
+    "launch_tools_mcp_server": _t("praisonaiagents.mcp.mcp_server", "launch_tools_mcp_server"),
+    "function_to_mcp_schema": _t("praisonaiagents.mcp.mcp_utils", "function_to_mcp_schema"),
+    "get_tool_metadata": _t("praisonaiagents.mcp.mcp_utils", "get_tool_metadata"),
+    "python_type_to_json_schema": _t("praisonaiagents.mcp.mcp_utils", "python_type_to_json_schema"),
+    "fix_array_schemas": _t("praisonaiagents.mcp.mcp_schema_utils", "fix_array_schemas"),
+    "ThreadLocalEventLoop": _t("praisonaiagents.mcp.mcp_schema_utils", "ThreadLocalEventLoop"),
+    "get_thread_local_event_loop": _t("praisonaiagents.mcp.mcp_schema_utils", "get_thread_local_event_loop"),
+    # Auth storage (lazy loaded) — import targets, not credentials.
+    "MCPAuthStorage": _t("praisonaiagents.mcp.mcp_auth_storage", "MCPAuthStorage"),
+    "get_default_auth_filepath": _t("praisonaiagents.mcp.mcp_auth_storage", "get_default_auth_filepath"),
+    # OAuth callback utilities (lazy loaded) — import targets, not credentials.
+    "OAuthCallbackHandler": _t("praisonaiagents.mcp.mcp_oauth_callback", "OAuthCallbackHandler"),
+    "generate_state": _t("praisonaiagents.mcp.mcp_oauth_callback", "generate_state"),
+    "generate_code_verifier": _t("praisonaiagents.mcp.mcp_oauth_callback", "generate_code_verifier"),
+    "generate_code_challenge": _t("praisonaiagents.mcp.mcp_oauth_callback", "generate_code_challenge"),
+    "get_redirect_url": _t("praisonaiagents.mcp.mcp_oauth_callback", "get_redirect_url"),
+    "OAUTH_CALLBACK_PORT": _t("praisonaiagents.mcp.mcp_oauth_callback", "OAUTH_CALLBACK_PORT"),
+    "OAUTH_CALLBACK_PATH": _t("praisonaiagents.mcp.mcp_oauth_callback", "OAUTH_CALLBACK_PATH"),
+    # Zero-config OAuth: discovery, dynamic registration, orchestration.
+    "discover_oauth_metadata": _t("praisonaiagents.mcp.oauth_discovery", "discover_oauth_metadata"),
+    "parse_www_authenticate": _t("praisonaiagents.mcp.oauth_discovery", "parse_www_authenticate"),
+    "register_client": _t("praisonaiagents.mcp.oauth_registration", "register_client"),
+    "MCPOAuthProvider": _t("praisonaiagents.mcp.oauth_provider", "MCPOAuthProvider"),
+    "InteractiveAuthRequired": _t("praisonaiagents.mcp.oauth_provider", "InteractiveAuthRequired"),
     # Protocols (lazy loaded)
-    elif name == "MCPClientProtocol":
-        from .protocols import MCPClientProtocol
-        return MCPClientProtocol
+    "MCPClientProtocol": _t("praisonaiagents.mcp.protocols", "MCPClientProtocol"),
     # Loader (lazy loaded)
-    elif name == "load_mcp_tools":
-        from .loader import load_mcp_tools
-        return load_mcp_tools
+    "load_mcp_tools": _t("praisonaiagents.mcp.loader", "load_mcp_tools"),
     # Resource/prompt normalisation helpers (lazy loaded)
-    elif name == "normalize_resource_result":
-        from .resources import normalize_resource_result
-        return normalize_resource_result
-    elif name == "normalize_prompt_result":
-        from .resources import normalize_prompt_result
-        return normalize_prompt_result
-    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
+    "normalize_resource_result": _t("praisonaiagents.mcp.resources", "normalize_resource_result"),
+    "normalize_prompt_result": _t("praisonaiagents.mcp.resources", "normalize_prompt_result"),
+}
+
+__getattr__ = create_lazy_getattr(_LAZY_IMPORTS, __name__)
 
 __all__ = [
     # Client
@@ -138,4 +106,10 @@ __all__ = [
     "get_redirect_url",
     "OAUTH_CALLBACK_PORT",
     "OAUTH_CALLBACK_PATH",
+    # Zero-config OAuth (lazy loaded)
+    "discover_oauth_metadata",
+    "parse_www_authenticate",
+    "register_client",
+    "MCPOAuthProvider",
+    "InteractiveAuthRequired",
 ]

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

@@ -428,9 +428,15 @@ class MCP:
                 self.is_npx = False
                 return
             
-        # Handle the single string format for stdio client
-        if isinstance(command_or_string, str) and args is None:
-            # Split the string into command and args using shell-like parsing
+        # Handle the single string format for stdio client.
+        # Shell-split the command string first, then decide based on how many
+        # tokens it yields. This makes multi-word forms like
+        # MCP("npx -y pkg", args=["/tmp"]) work — the first token is the
+        # executable and the remaining tokens are prepended to any explicitly
+        # supplied args. Using shlex handles all whitespace (spaces, tabs,
+        # newlines) and lets an executable path that contains spaces be passed
+        # intact by quoting it, e.g. MCP('"/opt/my app"', args=["run"]).
+        if isinstance(command_or_string, str):
             if platform.system() == 'Windows':
                 # Use shlex with posix=False for Windows to handle quotes and paths with spaces
                 parts = shlex.split(command_or_string, posix=False)
@@ -438,13 +444,19 @@ class MCP:
                 parts = [part.strip('"') for part in parts]
             else:
                 parts = shlex.split(command_or_string)
-            if not parts:
-                raise ValueError("Empty command string")
-            
-            cmd = parts[0]
-            arguments = parts[1:] if len(parts) > 1 else []
+
+            if len(parts) > 1:
+                # Multi-token string: first token is the executable, the rest
+                # are prepended to any explicitly supplied args.
+                cmd = parts[0]
+                arguments = parts[1:] + (args or [])
+            else:
+                # Single token (or a quoted spaced path): treat the whole
+                # string as a bare executable and keep explicit args as-is.
+                cmd = parts[0] if parts else command_or_string
+                arguments = args or []
         else:
-            # Use the original format with separate command and args
+            # Use the original format with a bare command and separate args
             cmd = command_or_string
             arguments = args or []
         

src/praisonai-agents/praisonaiagents/mcp/oauth_discovery.py

@@ -0,0 +1,231 @@
+"""
+MCP OAuth Metadata Discovery.
+
+Implements the discovery half of zero-config OAuth for remote MCP servers:
+
+- Parse a ``WWW-Authenticate: Bearer resource_metadata="..."`` challenge from a
+  ``401`` response (RFC 9728 / MCP authorization spec).
+- Fetch protected-resource metadata (``/.well-known/oauth-protected-resource``).
+- Fetch authorization-server metadata
+  (``/.well-known/oauth-authorization-server`` with an OpenID configuration
+  fallback), yielding ``authorization_endpoint``, ``token_endpoint`` and
+  ``registration_endpoint``.
+
+Network access is HTTPS-only (except loopback) to preserve the SSRF-safe
+posture used elsewhere in the MCP transports. ``requests`` is imported lazily so
+this module stays additive and does not affect import time.
+"""
+
+import re
+from typing import Any, Dict, List, Optional
+from urllib.parse import urljoin, urlparse
+
+from praisonaiagents._logging import get_logger
+
+logger = get_logger("mcp-oauth-discovery")
+
+_LOCALHOST_HOSTS = {"localhost", "127.0.0.1", "::1"}
+
+
+def _require_https(url: str) -> None:
+    """Reject non-HTTPS metadata URLs (loopback HTTP is allowed for testing)."""
+    parsed = urlparse(url)
+    if parsed.scheme == "https":
+        return
+    if parsed.scheme == "http" and (parsed.hostname or "") in _LOCALHOST_HOSTS:
+        return
+    raise ValueError(
+        f"Refusing to fetch OAuth metadata over insecure transport: {url!r}"
+    )
+
+
+def parse_www_authenticate(header: Optional[str]) -> Dict[str, str]:
+    """
+    Parse a ``WWW-Authenticate`` challenge header into its parameters.
+
+    Only ``Bearer`` challenges are relevant here. Returns a dict of the
+    ``key="value"`` auth-params (e.g. ``resource_metadata``, ``scope``,
+    ``error``). Returns an empty dict when the header is missing or not a
+    Bearer challenge.
+    """
+    if not header:
+        return {}
+
+    # Strip the leading scheme (e.g. "Bearer ") if present.
+    stripped = header.strip()
+    if stripped.lower().startswith("bearer"):
+        stripped = stripped[len("bearer"):].lstrip()
+
+    params: Dict[str, str] = {}
+    for match in re.finditer(r'(\w+)\s*=\s*"([^"]*)"', stripped):
+        params[match.group(1).lower()] = match.group(2)
+    return params
+
+
+def _get_json(url: str, timeout: float = 10.0) -> Dict[str, Any]:
+    """Fetch and parse a JSON metadata document over HTTPS.
+
+    Redirects are followed but every hop — including the final destination — is
+    re-validated with :func:`_require_https` so a redirect cannot escape the
+    HTTPS/loopback guard to an internal or plaintext endpoint (SSRF-safe).
+    """
+    _require_https(url)
+    import requests  # lazy import — optional dependency
+
+    resp = requests.get(
+        url,
+        headers={"Accept": "application/json"},
+        timeout=timeout,
+        allow_redirects=True,
+    )
+    # Re-validate every redirect hop and the final URL: allow_redirects only
+    # checks the initial URL, so a 30x to http:// or an internal host would
+    # otherwise bypass the guard.
+    for previous in getattr(resp, "history", None) or []:
+        _require_https(previous.url)
+    _require_https(resp.url)
+    resp.raise_for_status()
+    return resp.json()
+
+
+def _wellknown_candidates(server_url: str) -> List[str]:
+    """Build ``.well-known`` metadata URL candidates for a server URL."""
+    parsed = urlparse(server_url)
+    origin = f"{parsed.scheme}://{parsed.netloc}"
+    return [
+        urljoin(origin + "/", ".well-known/oauth-protected-resource"),
+        urljoin(origin + "/", ".well-known/oauth-authorization-server"),
+        urljoin(origin + "/", ".well-known/openid-configuration"),
+    ]
+
+
+def discover_protected_resource(
+    server_url: str,
+    resource_metadata_url: Optional[str] = None,
+    timeout: float = 10.0,
+) -> Optional[Dict[str, Any]]:
+    """
+    Discover protected-resource metadata for an MCP server.
+
+    Args:
+        server_url: The MCP server URL that returned a 401.
+        resource_metadata_url: Optional explicit URL from the
+            ``WWW-Authenticate: resource_metadata="..."`` challenge.
+        timeout: Per-request timeout in seconds.
+
+    Returns:
+        The protected-resource metadata document, or None if unavailable.
+    """
+    url = resource_metadata_url or _wellknown_candidates(server_url)[0]
+    try:
+        return _get_json(url, timeout=timeout)
+    except Exception as exc:  # noqa: BLE001 - discovery is best-effort
+        logger.debug(f"Protected-resource metadata unavailable at {url}: {exc}")
+        return None
+
+
+def discover_authorization_server(
+    server_url: str,
+    authorization_servers: Optional[List[str]] = None,
+    timeout: float = 10.0,
+) -> Optional[Dict[str, Any]]:
+    """
+    Discover authorization-server metadata.
+
+    Tries, in order: any ``authorization_servers`` advertised by the
+    protected-resource metadata, then the server-origin
+    ``/.well-known/oauth-authorization-server`` and OpenID-configuration
+    fallbacks.
+
+    Returns the first metadata document that exposes at least an
+    ``authorization_endpoint`` and ``token_endpoint``, or None.
+    """
+    candidates: List[str] = []
+
+    for issuer in authorization_servers or []:
+        issuer = issuer.rstrip("/")
+        candidates.append(issuer + "/.well-known/oauth-authorization-server")
+        candidates.append(issuer + "/.well-known/openid-configuration")
+
+    # Fall back to the resource server's own origin.
+    for candidate in _wellknown_candidates(server_url)[1:]:
+        candidates.append(candidate)
+
+    for url in candidates:
+        try:
+            metadata = _get_json(url, timeout=timeout)
+        except Exception as exc:  # noqa: BLE001 - best-effort discovery
+            logger.debug(f"Auth-server metadata unavailable at {url}: {exc}")
+            continue
+        if metadata.get("authorization_endpoint") and metadata.get("token_endpoint"):
+            return metadata
+
+    return None
+
+
+def discover_oauth_metadata(
+    server_url: str,
+    www_authenticate: Optional[str] = None,
+    timeout: float = 10.0,
+) -> Optional[Dict[str, Any]]:
+    """
+    Full metadata discovery for an OAuth-protected MCP server.
+
+    Combines the ``WWW-Authenticate`` challenge (if any), protected-resource
+    metadata, and authorization-server metadata into a single normalised dict:
+
+        {
+            "authorization_endpoint": str,
+            "token_endpoint": str,
+            "registration_endpoint": str | None,
+            "scopes_supported": list[str] | None,
+            "resource": str | None,
+        }
+
+    Returns None if a usable authorization/token endpoint pair cannot be found.
+    """
+    challenge = parse_www_authenticate(www_authenticate)
+
+    resource_meta = discover_protected_resource(
+        server_url,
+        resource_metadata_url=challenge.get("resource_metadata"),
+        timeout=timeout,
+    )
+
+    authorization_servers = None
+    resource_id = None
+    if resource_meta:
+        authorization_servers = resource_meta.get("authorization_servers")
+        resource_id = resource_meta.get("resource")
+
+    auth_meta = discover_authorization_server(
+        server_url,
+        authorization_servers=authorization_servers,
+        timeout=timeout,
+    )
+    if not auth_meta:
+        return None
+
+    # Enforce the HTTPS/loopback guard on the advertised endpoints too: an
+    # HTTPS metadata document must not be able to smuggle a plaintext
+    # authorization/token/registration endpoint that would later receive
+    # authorization codes, PKCE verifiers, refresh tokens, or client secrets
+    # over an insecure transport.
+    authorization_endpoint = auth_meta["authorization_endpoint"]
+    token_endpoint = auth_meta["token_endpoint"]
+    registration_endpoint = auth_meta.get("registration_endpoint")
+    _require_https(authorization_endpoint)
+    _require_https(token_endpoint)
+    if registration_endpoint:
+        _require_https(registration_endpoint)
+
+    return {
+        "authorization_endpoint": authorization_endpoint,
+        "token_endpoint": token_endpoint,
+        "registration_endpoint": registration_endpoint,
+        "scopes_supported": (
+            auth_meta.get("scopes_supported")
+            or (resource_meta or {}).get("scopes_supported")
+        ),
+        "resource": resource_id,
+    }

src/praisonai-agents/praisonaiagents/mcp/oauth_provider.py

@@ -0,0 +1,340 @@
+"""
+MCP OAuth Provider — zero-config orchestration.
+
+Composes the OAuth building blocks into a single ``ensure_authenticated`` entry
+point that transports invoke on a ``401``:
+
+    discovery  ->  dynamic client registration (if no client stored)
+               ->  PKCE authorization-code flow (browser + loopback callback)
+               ->  token persistence
+               ->  transparent refresh on expiry
+
+All heavy work (network I/O, browser, loopback HTTP server) is lazily imported
+and only runs when a server actually challenges for OAuth, so servers without
+auth — or with pre-supplied client info/tokens — behave exactly as before.
+"""
+
+import time
+import webbrowser
+from http.server import BaseHTTPRequestHandler, HTTPServer
+from threading import Thread
+from typing import Any, Dict, Optional
+from urllib.parse import urlencode, urlparse, parse_qs
+
+from praisonaiagents._logging import get_logger
+
+from .mcp_oauth_callback import (
+    OAuthCallbackHandler,
+    OAUTH_CALLBACK_PORT,
+    OAUTH_CALLBACK_PATH,
+    generate_state,
+    generate_code_verifier,
+    generate_code_challenge,
+    get_redirect_url,
+)
+
+logger = get_logger("mcp-oauth-provider")
+
+
+class InteractiveAuthRequired(RuntimeError):
+    """Raised when interactive OAuth is needed but not possible (headless/CI)."""
+
+
+def _make_callback_server(handler: OAuthCallbackHandler, port: int) -> HTTPServer:
+    """Build a minimal loopback HTTP server that feeds codes to ``handler``."""
+
+    class _Handler(BaseHTTPRequestHandler):
+        def do_GET(self):  # noqa: N802 - required name
+            parsed = urlparse(self.path)
+            if parsed.path != OAUTH_CALLBACK_PATH:
+                self.send_response(404)
+                self.end_headers()
+                return
+            params = parse_qs(parsed.query)
+            state = (params.get("state") or [""])[0]
+            code = (params.get("code") or [""])[0]
+            if state and code:
+                handler.receive_callback(state, code)
+                body = b"Authentication complete. You can close this window."
+            else:
+                body = b"Authentication failed: missing code or state."
+            self.send_response(200)
+            self.send_header("Content-Type", "text/plain")
+            self.end_headers()
+            self.wfile.write(body)
+
+        def log_message(self, *args, **kwargs):  # silence default logging
+            pass
+
+    return HTTPServer(("127.0.0.1", port), _Handler)
+
+
+class MCPOAuthProvider:
+    """
+    Orchestrates zero-config OAuth for a single remote MCP server.
+
+    Args:
+        mcp_name: Stable identifier used as the storage key.
+        server_url: The MCP server URL.
+        storage: Optional pre-built ``MCPAuthStorage`` (created lazily otherwise).
+        open_browser: Whether to open a browser for interactive auth. Set False
+            for headless/CI to get a single actionable ``InteractiveAuthRequired``.
+    """
+
+    def __init__(
+        self,
+        mcp_name: str,
+        server_url: str,
+        storage: Optional[Any] = None,
+        open_browser: bool = True,
+    ):
+        self.mcp_name = mcp_name
+        self.server_url = server_url
+        self.open_browser = open_browser
+        if storage is None:
+            from .mcp_auth_storage import MCPAuthStorage
+            storage = MCPAuthStorage()
+        self.storage = storage
+
+    def get_valid_token(self) -> Optional[str]:
+        """Return a non-expired access token if one is stored, else None."""
+        entry = self.storage.get_for_url(self.mcp_name, self.server_url)
+        if not entry or not entry.get("tokens"):
+            return None
+        if self.storage.is_token_expired(self.mcp_name):
+            refreshed = self._try_refresh(entry)
+            if not refreshed:
+                return None
+            entry = self.storage.get_for_url(self.mcp_name, self.server_url) or {}
+        tokens = entry.get("tokens") or {}
+        return tokens.get("access_token")
+
+    def ensure_authenticated(
+        self,
+        www_authenticate: Optional[str] = None,
+        timeout: float = 10.0,
+    ) -> str:
+        """
+        Ensure a valid access token exists, running the full flow if needed.
+
+        Returns:
+            A valid bearer access token.
+
+        Raises:
+            InteractiveAuthRequired: If interactive auth is needed but disabled.
+            RuntimeError: If discovery fails or the flow cannot complete.
+        """
+        token = self.get_valid_token()
+        if token:
+            return token
+
+        from .oauth_discovery import discover_oauth_metadata
+
+        metadata = discover_oauth_metadata(
+            self.server_url,
+            www_authenticate=www_authenticate,
+            timeout=timeout,
+        )
+        if not metadata:
+            raise RuntimeError(
+                f"Could not discover OAuth metadata for {self.server_url}. "
+                "The server may not advertise standard authorization-server "
+                "metadata; supply client credentials manually."
+            )
+
+        client_info = self._ensure_client(metadata, timeout=timeout)
+
+        scope = None
+        scopes = metadata.get("scopes_supported")
+        if scopes:
+            scope = " ".join(scopes)
+
+        return self._run_pkce_flow(metadata, client_info, scope, timeout=timeout)
+
+    def _ensure_client(
+        self,
+        metadata: Dict[str, Any],
+        timeout: float,
+    ) -> Dict[str, Any]:
+        """Return stored client info or perform dynamic client registration."""
+        entry = self.storage.get_for_url(self.mcp_name, self.server_url)
+        if entry and entry.get("client_info", {}).get("client_id"):
+            return entry["client_info"]
+
+        registration_endpoint = metadata.get("registration_endpoint")
+        if not registration_endpoint:
+            raise RuntimeError(
+                f"No client is registered for {self.server_url} and the "
+                "authorization server does not support dynamic client "
+                "registration (no registration_endpoint)."
+            )
+
+        from .oauth_registration import register_client
+
+        scope = None
+        scopes = metadata.get("scopes_supported")
+        if scopes:
+            scope = " ".join(scopes)
+
+        client_info = register_client(
+            registration_endpoint,
+            redirect_uris=[get_redirect_url()],
+            scope=scope,
+            timeout=timeout,
+        )
+        self.storage.set_client_info(
+            self.mcp_name, client_info, server_url=self.server_url
+        )
+        return client_info
+
+    def _run_pkce_flow(
+        self,
+        metadata: Dict[str, Any],
+        client_info: Dict[str, Any],
+        scope: Optional[str],
+        timeout: float,
+    ) -> str:
+        """Run the interactive PKCE authorization-code flow and store tokens."""
+        if not self.open_browser:
+            raise InteractiveAuthRequired(
+                f"Interactive OAuth is required to authenticate with "
+                f"{self.server_url}. Run this from an interactive session, or "
+                f"pre-supply tokens, to complete authentication."
+            )
+
+        callback = OAuthCallbackHandler()
+        state = generate_state()
+        code_verifier = generate_code_verifier()
+        code_challenge = generate_code_challenge(code_verifier)
+
+        params = {
+            "response_type": "code",
+            "client_id": client_info["client_id"],
+            "redirect_uri": get_redirect_url(),
+            "state": state,
+            "code_challenge": code_challenge,
+            "code_challenge_method": "S256",
+        }
+        if scope:
+            params["scope"] = scope
+        if metadata.get("resource"):
+            params["resource"] = metadata["resource"]
+
+        auth_url = f"{metadata['authorization_endpoint']}?{urlencode(params)}"
+
+        server = _make_callback_server(callback, OAUTH_CALLBACK_PORT)
+        thread = Thread(target=server.serve_forever, daemon=True)
+        thread.start()
+        try:
+            logger.debug(f"Opening browser for OAuth authorization: {auth_url}")
+            webbrowser.open(auth_url)
+            code = callback.wait_for_callback(state, timeout=300.0)
+        finally:
+            server.shutdown()
+
+        tokens = self._exchange_code(
+            metadata, client_info, code, code_verifier, timeout=timeout
+        )
+        self.storage.set_tokens(
+            self.mcp_name, tokens, server_url=self.server_url
+        )
+        return tokens["access_token"]
+
+    def _exchange_code(
+        self,
+        metadata: Dict[str, Any],
+        client_info: Dict[str, Any],
+        code: str,
+        code_verifier: str,
+        timeout: float,
+    ) -> Dict[str, Any]:
+        """Exchange an authorization code for tokens at the token endpoint."""
+        import requests  # lazy import — optional dependency
+
+        data = {
+            "grant_type": "authorization_code",
+            "code": code,
+            "redirect_uri": get_redirect_url(),
+            "client_id": client_info["client_id"],
+            "code_verifier": code_verifier,
+        }
+        if client_info.get("client_secret"):
+            data["client_secret"] = client_info["client_secret"]
+        if metadata.get("resource"):
+            data["resource"] = metadata["resource"]
+
+        resp = requests.post(
+            metadata["token_endpoint"],
+            data=data,
+            headers={"Accept": "application/json"},
+            timeout=timeout,
+        )
+        resp.raise_for_status()
+        return self._normalize_tokens(
+            resp.json(), token_endpoint=metadata["token_endpoint"]
+        )
+
+    def _try_refresh(self, entry: Dict[str, Any]) -> bool:
+        """Attempt a refresh-token grant; return True on success."""
+        tokens = entry.get("tokens") or {}
+        refresh_token = tokens.get("refresh_token")
+        token_endpoint = tokens.get("token_endpoint")
+        client_info = entry.get("client_info") or {}
+        client_id = client_info.get("client_id")
+        if not (refresh_token and token_endpoint and client_id):
+            return False
+
+        try:
+            from .oauth_discovery import _require_https
+
+            # Defense in depth: never POST a refresh token / client secret to a
+            # persisted endpoint that isn't HTTPS (or loopback).
+            _require_https(token_endpoint)
+
+            import requests  # lazy import — optional dependency
+
+            data = {
+                "grant_type": "refresh_token",
+                "refresh_token": refresh_token,
+                "client_id": client_id,
+            }
+            if client_info.get("client_secret"):
+                data["client_secret"] = client_info["client_secret"]
+
+            resp = requests.post(
+                token_endpoint,
+                data=data,
+                headers={"Accept": "application/json"},
+                timeout=10.0,
+            )
+            resp.raise_for_status()
+            new_tokens = self._normalize_tokens(
+                resp.json(), token_endpoint=token_endpoint
+            )
+            # Preserve refresh_token if the server didn't return a new one.
+            if not new_tokens.get("refresh_token"):
+                new_tokens["refresh_token"] = refresh_token
+            self.storage.set_tokens(
+                self.mcp_name, new_tokens, server_url=self.server_url
+            )
+            return True
+        except Exception as exc:  # noqa: BLE001 - refresh is best-effort
+            logger.debug(f"Token refresh failed for {self.mcp_name}: {exc}")
+            return False
+
+    @staticmethod
+    def _normalize_tokens(
+        raw: Dict[str, Any],
+        token_endpoint: Optional[str] = None,
+    ) -> Dict[str, Any]:
+        """Normalise a token response, computing an absolute ``expires_at``."""
+        tokens = dict(raw)
+        expires_in = tokens.get("expires_in")
+        if expires_in and not tokens.get("expires_at"):
+            try:
+                tokens["expires_at"] = time.time() + float(expires_in)
+            except (TypeError, ValueError):
+                pass
+        if token_endpoint:
+            tokens["token_endpoint"] = token_endpoint
+        return tokens

src/praisonai-agents/praisonaiagents/mcp/oauth_registration.py

@@ -0,0 +1,91 @@
+"""
+MCP OAuth Dynamic Client Registration (RFC 7591).
+
+Performs an anonymous dynamic client registration against an authorization
+server's ``registration_endpoint`` so that connecting to a hosted MCP server by
+URL requires no pre-provisioned ``client_id``. The registered client info is
+returned for persistence via :meth:`MCPAuthStorage.set_client_info`.
+
+``requests`` is imported lazily to keep this module additive.
+"""
+
+from typing import Any, Dict, List, Optional
+from urllib.parse import urlparse
+
+from praisonaiagents._logging import get_logger
+
+logger = get_logger("mcp-oauth-registration")
+
+_LOCALHOST_HOSTS = {"localhost", "127.0.0.1", "::1"}
+
+
+def _require_https(url: str) -> None:
+    """Reject non-HTTPS registration URLs (loopback HTTP allowed for testing)."""
+    parsed = urlparse(url)
+    if parsed.scheme == "https":
+        return
+    if parsed.scheme == "http" and (parsed.hostname or "") in _LOCALHOST_HOSTS:
+        return
+    raise ValueError(
+        f"Refusing to register client over insecure transport: {url!r}"
+    )
+
+
+def register_client(
+    registration_endpoint: str,
+    redirect_uris: List[str],
+    client_name: str = "PraisonAI Agent",
+    scope: Optional[str] = None,
+    timeout: float = 10.0,
+) -> Dict[str, Any]:
+    """
+    Register a public OAuth client via RFC 7591 Dynamic Client Registration.
+
+    Args:
+        registration_endpoint: The ``registration_endpoint`` from auth-server
+            metadata.
+        redirect_uris: Allowed redirect URIs (the loopback callback URL).
+        client_name: Human-readable client name.
+        scope: Optional space-delimited scope string to request.
+        timeout: Request timeout in seconds.
+
+    Returns:
+        The registration response containing at least ``client_id`` (and
+        ``client_secret`` for confidential clients).
+
+    Raises:
+        ValueError: If the endpoint is insecure or the response lacks a
+            ``client_id``.
+    """
+    _require_https(registration_endpoint)
+    import requests  # lazy import — optional dependency
+
+    payload: Dict[str, Any] = {
+        "client_name": client_name,
+        "redirect_uris": redirect_uris,
+        "grant_types": ["authorization_code", "refresh_token"],
+        "response_types": ["code"],
+        "token_endpoint_auth_method": "none",
+    }
+    if scope:
+        payload["scope"] = scope
+
+    resp = requests.post(
+        registration_endpoint,
+        json=payload,
+        headers={"Content-Type": "application/json", "Accept": "application/json"},
+        timeout=timeout,
+    )
+    resp.raise_for_status()
+    client_info = resp.json()
+
+    if not client_info.get("client_id"):
+        raise ValueError(
+            "Dynamic client registration response missing 'client_id'"
+        )
+
+    logger.debug(
+        f"Registered OAuth client '{client_info.get('client_id')}' "
+        f"at {registration_endpoint}"
+    )
+    return client_info

src/praisonai-agents/praisonaiagents/memory/adapters/__init__.py

@@ -26,6 +26,7 @@ from .registry import (
     list_memory_adapters,
     get_first_available_memory_adapter,
     has_memory_adapter,
+    resolve_memory_adapter_name,
 )
 from .factories import (
     create_mem0_memory_adapter,
@@ -53,4 +54,5 @@ __all__ = [
     'list_memory_adapters',
     'get_first_available_memory_adapter',
     'has_memory_adapter',
+    'resolve_memory_adapter_name',
 ]

src/praisonai-agents/praisonaiagents/memory/adapters/factories.py

@@ -371,6 +371,39 @@ class ChromaMemoryAdapter:
                 sanitized[k] = str(v)
         return sanitized
     
+    def _reset_collection(self) -> None:
+        """Delete and recreate this adapter's collection, wiping stored memory.
+
+        Short-term and long-term share one collection in this adapter, so both
+        resets route here. Scoped to this adapter's own collection to avoid
+        wiping other collections in the shared persistent store.
+        """
+        try:
+            self.client.delete_collection(name=self.collection_name)
+        except Exception:
+            pass
+        self.collection = self.client.create_collection(
+            name=self.collection_name,
+            metadata={"hnsw:space": "cosine"}
+        )
+
+    def reset_short_term(self) -> None:
+        """Clear this adapter's memory.
+
+        NOTE: this adapter stores short-term and long-term in a single shared
+        collection (``store_short_term``/``search_short_term`` delegate to their
+        long-term counterparts and no ``memory_type`` tag is recorded), so
+        per-tier scoping is not possible here — resetting short-term also clears
+        long-term. Use a tier-aware backend (e.g. the Dakera adapter) if you need
+        to erase a single tier.
+        """
+        self._reset_collection()
+
+    def reset_long_term(self) -> None:
+        """Clear this adapter's memory (shared with short-term; see
+        ``reset_short_term`` for the shared-collection caveat)."""
+        self._reset_collection()
+
     def close(self):
         """Clean up ChromaDB resources."""
         # ChromaDB PersistentClient has no explicit close; release references.

src/praisonai-agents/praisonaiagents/memory/adapters/registry.py

@@ -77,6 +77,17 @@ def has_memory_adapter(name: str) -> bool:
     return get_default_memory_registry().is_available(name)
 
 
+#: Accepted spellings for a backend, mapped to the adapter that serves them.
+#: Single source of truth for MemoryConfig, Agent and Memory.
+MEMORY_PROVIDER_ALIASES = {"rag": "chroma", "chromadb": "chroma", "none": "in_memory"}
+
+
+def resolve_memory_adapter_name(name: str) -> Optional[str]:
+    """Adapter name for *name*, or None if nothing is registered under it."""
+    candidate = MEMORY_PROVIDER_ALIASES.get(name, name)
+    return candidate if has_memory_adapter(candidate) else None
+
+
 def get_first_available_memory_adapter(
     preferences: Optional[List[str]] = None, **kwargs
 ) -> Optional[Tuple[str, MemoryProtocol]]:
@@ -107,4 +118,6 @@ __all__ = [
     'add_memory_adapter',
     'add_memory_factory',
     'has_memory_adapter',
+    'resolve_memory_adapter_name',
+    'MEMORY_PROVIDER_ALIASES',
 ]

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

@@ -136,16 +136,21 @@ class Memory(SearchMixin, MemoryCoreMixin):
         provider = self.cfg.get("provider", "rag").lower()
         self._log_verbose(f"Requested memory provider: {provider}")
         
-        # Map legacy provider names to adapter names
-        provider_mapping = {
-            "rag": "chroma",  # Legacy "rag" provider uses ChromaDB
-            "mem0": "mem0",
-            "mongodb": "mongodb",
-            "sqlite": "sqlite",
-            "none": "in_memory"
-        }
-        
-        adapter_name = provider_mapping.get(provider, provider)
+        # Legacy provider aliases live in memory/adapters/registry.py
+        from .adapters.registry import (
+            MEMORY_PROVIDER_ALIASES, list_memory_adapters, resolve_memory_adapter_name)
+
+        # A provider with no adapter must fail loudly: falling through landed
+        # on the SQLite fallback below, so provider="redis" silently produced a
+        # local .db file per process.
+        adapter_name = resolve_memory_adapter_name(provider)
+        if adapter_name is None:
+            from ..config.parse_utils import make_preset_error
+            raise make_preset_error(
+                "memory provider",
+                provider,
+                sorted(set(list_memory_adapters()) | set(MEMORY_PROVIDER_ALIASES)),
+            )
         
         # Try to get preferred adapter, fallback to available ones
         adapter = None
@@ -195,8 +200,15 @@ class Memory(SearchMixin, MemoryCoreMixin):
                 self.provider = adapter_name
                 return
                 
-            # Fallback to first available adapter
-            self._log_verbose(f"Provider '{adapter_name}' not available, trying fallbacks")
+            # Falling back changes where the data lives - say so out loud when
+            # the user named a specific provider.
+            _fallback_log = (
+                logger.warning if provider_explicitly_requested else self._log_verbose
+            )
+            _fallback_log(
+                f"Memory provider '{adapter_name}' is not available; falling back "
+                f"to a local store. Data will NOT be in '{adapter_name}'."
+            )
             # Try each fallback preference individually
             for fallback_provider in ["sqlite", "in_memory"]:
                 try:
@@ -204,7 +216,7 @@ class Memory(SearchMixin, MemoryCoreMixin):
                     adapter = get_memory_adapter(fallback_provider, **fallback_config)
                     if adapter:
                         adapter_name = fallback_provider
-                        self._log_verbose(f"Using fallback adapter: {adapter_name}")
+                        _fallback_log(f"Using fallback memory adapter: {adapter_name}")
                         break
                 except Exception as e:
                     self._log_verbose(f"Fallback {fallback_provider} failed: {e}")
@@ -814,6 +826,11 @@ class Memory(SearchMixin, MemoryCoreMixin):
         if getattr(self, "memory_adapter", None):
             if hasattr(self.memory_adapter, "reset_short_term"):
                 self.memory_adapter.reset_short_term()
+            else:
+                logger.warning(
+                    f"{type(self.memory_adapter).__name__} does not support "
+                    "reset_short_term(); short-term memory was NOT cleared"
+                )
             return
         conn = self._get_stm_conn()
         with self._write_lock:  # Serialize write operations
@@ -1199,6 +1216,11 @@ class Memory(SearchMixin, MemoryCoreMixin):
         if getattr(self, "memory_adapter", None):
             if hasattr(self.memory_adapter, "reset_long_term"):
                 self.memory_adapter.reset_long_term()
+            else:
+                logger.warning(
+                    f"{type(self.memory_adapter).__name__} does not support "
+                    "reset_long_term(); long-term memory was NOT cleared"
+                )
             return
         conn = self._get_ltm_conn()
         with self._write_lock:  # Serialize write operations

src/praisonai-agents/praisonaiagents/output/models.py

@@ -6,11 +6,16 @@ It is kept separate from main.py to avoid importing rich at module level.
 """
 
 import json
-from typing import Optional, Dict, Any, Literal, TYPE_CHECKING
+from typing import Optional, Dict, Any, Literal
 from pydantic import BaseModel, ConfigDict
 
-if TYPE_CHECKING:
+# Import token metrics at runtime so the forward reference in TaskOutput can be
+# resolved by model_rebuild() below. Importing only under TYPE_CHECKING left the
+# model perpetually "not fully defined", making any instantiation raise.
+try:
     from ..telemetry.token_collector import TokenMetrics
+except ImportError:
+    TokenMetrics = None
 
 
 class TaskOutput(BaseModel):
@@ -52,3 +57,8 @@ class ReflectionOutput(BaseModel):
     """Output model for self-reflection results."""
     reflection: str
     satisfactory: Literal["yes", "no"]
+
+
+# Resolve the 'TokenMetrics' forward reference now that it is available at
+# runtime, so TaskOutput can actually be instantiated.
+TaskOutput.model_rebuild()

src/praisonai-agents/praisonaiagents/paths.py

@@ -1,8 +1,18 @@
 """
 Centralized Path Utilities for PraisonAI Agents.
 
-All persistent data uses ~/.praisonai/ by default.
-Override with PRAISONAI_HOME environment variable.
+All persistent data uses ~/.praisonai/ by default. Override with the
+PRAISONAI_HOME environment variable for a single-root layout.
+
+For fresh installs (no PRAISONAI_HOME and no existing ~/.praisonai or
+~/.praison), the XDG Base Directory specification is honoured when the
+corresponding variables are set:
+    config -> $XDG_CONFIG_HOME/praisonai  (else ~/.config/praisonai)
+    data   -> $XDG_DATA_HOME/praisonai    (else ~/.praisonai)
+    state  -> $XDG_STATE_HOME/praisonai   (else ~/.local/state/praisonai)
+    cache  -> $XDG_CACHE_HOME/praisonai   (else ~/.cache/praisonai)
+An existing single root (PRAISONAI_HOME or ~/.praisonai/~/.praison) keeps all
+classes together, preserving full backward compatibility.
 
 This module provides a single source of truth for all data storage paths,
 eliminating hardcoded paths throughout the codebase (DRY principle).
@@ -40,14 +50,77 @@ DEFAULT_DIR_NAME = ".praisonai"
 # Legacy directory name (for backward compat)
 LEGACY_DIR_NAME = ".praison"
 
+# App directory name used under XDG base directories.
+XDG_APP_NAME = "praisonai"
+
 # Cache for data dir to avoid repeated filesystem checks
 _data_dir_cache: Optional[Path] = None
 
+# Cache for the single-root decision. Snapshotted on first resolution so that
+# creating the default data directory later in the same process does not
+# retroactively flip config/state/cache from their XDG locations onto a
+# single root (see Issue #3981). ``False`` means "resolved: no single root".
+_single_root_cache: Union[Path, bool, None] = None
+
 
 def _clear_cache() -> None:
-    """Clear the data dir cache. Used for testing."""
-    global _data_dir_cache
+    """Clear cached path decisions. Used for testing."""
+    global _data_dir_cache, _single_root_cache
     _data_dir_cache = None
+    _single_root_cache = None
+
+
+def _single_root() -> Optional[Path]:
+    """Return an explicit single-root home if one applies, else ``None``.
+
+    A single root applies when the user has opted into one (``PRAISONAI_HOME``)
+    or already has an existing legacy home (``~/.praisonai`` or ``~/.praison``).
+    In that case every directory class (config/data/state/cache) stays under
+    that single root, preserving full backward compatibility. When no such root
+    exists (a fresh install), ``None`` is returned so callers fall through to
+    the XDG base directories.
+
+    ``PRAISONAI_HOME`` is always honoured live (never cached) so tests and
+    runtime overrides take effect immediately. The existence-based legacy
+    detection, however, is snapshotted on first use: once a run has resolved
+    "no single root", later creation of the default ``~/.praisonai`` data
+    directory must not retroactively move config/state/cache onto it mid-run.
+    """
+    env_path = os.environ.get(ENV_VAR)
+    if env_path:
+        return Path(env_path).expanduser()
+
+    global _single_root_cache
+    if _single_root_cache is not None:
+        return _single_root_cache or None
+
+    home = Path.home()
+    new_path = home / DEFAULT_DIR_NAME
+    if new_path.exists():
+        _single_root_cache = new_path
+        return new_path
+
+    legacy_path = home / LEGACY_DIR_NAME
+    if legacy_path.exists():
+        _single_root_cache = legacy_path
+        return legacy_path
+
+    _single_root_cache = False
+    return None
+
+
+def _xdg_dir(base_env: str, default_subpath: str) -> Path:
+    """Resolve an XDG base directory for PraisonAI.
+
+    Uses ``$base_env`` when it is set to an absolute path, otherwise the XDG
+    default under the home directory, and appends the app name.
+    """
+    base = os.environ.get(base_env)
+    if base and os.path.isabs(base):
+        root = Path(base)
+    else:
+        root = Path.home() / default_subpath
+    return root / XDG_APP_NAME
 
 
 def get_data_dir() -> Path:
@@ -56,8 +129,11 @@ def get_data_dir() -> Path:
     
     Priority:
     1. PRAISONAI_HOME env var
-    2. ~/.praisonai/ (default)
+    2. ~/.praisonai/ (existing install)
     3. ~/.praison/ (legacy fallback with warning)
+    4. $XDG_DATA_HOME/praisonai (only when XDG_DATA_HOME is explicitly set
+       and no legacy root exists)
+    5. ~/.praisonai/ (default for fresh installs)
     
     Returns:
         Path to data directory
@@ -102,11 +178,53 @@ def get_data_dir() -> Path:
         _data_dir_cache = legacy_path
         return legacy_path
     
-    # Default to new location (will be created when needed)
+    # No single-root install present: honour XDG_DATA_HOME when explicitly set
+    # (servers/containers/CI point it at the correct persistent volume).
+    xdg_data = os.environ.get("XDG_DATA_HOME")
+    if xdg_data and os.path.isabs(xdg_data):
+        xdg_path = Path(xdg_data) / XDG_APP_NAME
+        _data_dir_cache = xdg_path
+        return xdg_path
+    
+    # Default to branded location (will be created when needed)
     _data_dir_cache = new_path
     return new_path
 
 
+def get_config_dir() -> Path:
+    """
+    Get PraisonAI config directory (editable config, credentials, rules).
+
+    Precedence:
+    1. ``PRAISONAI_HOME`` / existing legacy home (single root, back-compat)
+    2. ``$XDG_CONFIG_HOME/praisonai`` (else ``~/.config/praisonai``)
+
+    Returns:
+        Path to config directory
+    """
+    root = _single_root()
+    if root is not None:
+        return root
+    return _xdg_dir("XDG_CONFIG_HOME", ".config")
+
+
+def get_state_dir() -> Path:
+    """
+    Get PraisonAI state directory (machine-local: MRU model, logs, spill).
+
+    Precedence:
+    1. ``PRAISONAI_HOME`` / existing legacy home (single root, back-compat)
+    2. ``$XDG_STATE_HOME/praisonai`` (else ``~/.local/state/praisonai``)
+
+    Returns:
+        Path to state directory
+    """
+    root = _single_root()
+    if root is not None:
+        return root
+    return _xdg_dir("XDG_STATE_HOME", ".local/state")
+
+
 def get_sessions_dir() -> Path:
     """
     Get sessions directory.
@@ -126,9 +244,9 @@ def get_session_spill_dir() -> Path:
     next load (Issue #3597).
 
     Returns:
-        Path to ~/.praisonai/state/session_spill/
+        Path to <state>/state/session_spill/
     """
-    return get_data_dir() / "state" / "session_spill"
+    return get_state_dir() / "state" / "session_spill"
 
 
 def get_skills_dir() -> Path:
@@ -234,11 +352,17 @@ def get_learn_dir() -> Path:
 def get_cache_dir() -> Path:
     """
     Get cache directory (disposable data).
-    
+
+    Under a single-root install this is ``<root>/cache``; on a fresh install it
+    resolves to ``$XDG_CACHE_HOME/praisonai`` (else ``~/.cache/praisonai``).
+
     Returns:
-        Path to ~/.praisonai/cache/
+        Path to cache directory
     """
-    return get_data_dir() / "cache"
+    root = _single_root()
+    if root is not None:
+        return root / "cache"
+    return _xdg_dir("XDG_CACHE_HOME", ".cache")
 
 
 def get_mcp_auth_path() -> Path:
@@ -383,9 +507,9 @@ def get_config_path() -> Path:
     Get path to the main config.yaml file.
 
     Returns:
-        Path to ~/.praisonai/config.yaml
+        Path to <config>/config.yaml
     """
-    return get_data_dir() / "config.yaml"
+    return get_config_dir() / "config.yaml"
 
 
 def get_schedules_dir() -> Path:
@@ -443,6 +567,8 @@ def get_all_paths() -> Dict[str, Path]:
     """
     return {
         "data_dir": get_data_dir(),
+        "config_dir": get_config_dir(),
+        "state_dir": get_state_dir(),
         "sessions": get_sessions_dir(),
         "skills": get_skills_dir(),
         "plugins": get_plugins_dir(),

src/praisonai-agents/praisonaiagents/plugins/manager.py

@@ -828,7 +828,16 @@ def _adapt_plugin_hooks(plugin: Plugin) -> Iterator[Tuple["HookEvent", Callable]
 
     if _overrides("after_tool"):
         def after_tool_hook(data, _p=plugin):
-            _p.after_tool(getattr(data, "tool_name", ""), getattr(data, "tool_output", None))
+            try:
+                new_result = _p.after_tool(getattr(data, "tool_name", ""),
+                                           getattr(data, "tool_output", None))
+            except GuardrailBlocked as e:
+                return HookResult.block(e.reason)
+            decision = _as_decision(new_result)
+            if decision is not None:
+                return decision
+            if new_result is not None and hasattr(data, "tool_output"):
+                data.tool_output = new_result
             return HookResult.allow()
         yield HookEvent.AFTER_TOOL, after_tool_hook
 

src/praisonai-agents/praisonaiagents/plugins/plugin.py

@@ -26,9 +26,12 @@ class GuardrailBlocked(Exception):
     A ``POLICY``/``GUARDRAIL`` plugin can raise this from any ``before_*``
     method (``before_tool``, ``before_llm``, ``before_agent``,
     ``before_message``) to stop the tool call / LLM request / agent run /
-    inbound message. The plugin bridge converts it into a denying
-    ``HookResult`` so the runtime's existing ``is_blocked`` enforcement skips
-    the action and surfaces ``reason``.
+    inbound message. It may also be raised from ``after_tool`` to block
+    *propagation* of a tool result (e.g. secret/PII detected) — the tool has
+    already run, but the output is suppressed before it reaches the model.
+    The plugin bridge converts it into a denying ``HookResult`` so the
+    runtime's existing ``is_blocked`` enforcement skips the action and
+    surfaces ``reason``.
     """
 
     def __init__(self, reason: str = "Blocked by guardrail plugin"):
@@ -177,7 +180,13 @@ class Plugin(ABC):
         return args
     
     def after_tool(self, tool_name: str, result: Any) -> Any:
-        """Called after tool execution. Can modify result."""
+        """Called after tool execution.
+
+        Return a modified ``result`` (rewrite/redact) to replace the effective
+        tool output before it reaches the model or a channel, a deny/block
+        decision (``PluginDecision.deny(reason)`` / a ``HookResult`` / raise
+        :class:`GuardrailBlocked`), or ``None`` for no-op.
+        """
         return result
 
     def before_tool_definitions(

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

@@ -74,6 +74,10 @@ class Process:
                     self._state_lock = asyncio.Lock()
         return self._state_lock
 
+    def _task_by_name(self, name):
+        """Return the task whose name matches, or None. First-match, mirrors prior inline lookup."""
+        return next((t for t in self.tasks.values() if t.name == name), None)
+
     def _create_llm_instance(self):
         """Create and return a configured LLM instance for manager tasks."""
         from ..llm import LLM
@@ -115,7 +119,7 @@ class Process:
             if current_task.previous_tasks:
                 # For validation tasks, typically validate the most recent previous task
                 prev_task_name = current_task.previous_tasks[-1]
-                validated_task = next((t for t in self.tasks.values() if t.name == prev_task_name), None)
+                validated_task = self._task_by_name(prev_task_name)
             elif current_task.context:
                 # If no previous_tasks, check context for the validated task
                 for ctx_task in reversed(current_task.context):
@@ -348,6 +352,43 @@ class Process:
                     task_id=loop_task.id
                 )
 
+    def _build_routing_context(self, current_task: Task) -> Dict[str, Any]:
+        """Build a variable context for evaluating a task's `when` condition.
+
+        Combines the task's structured result (json_dict / pydantic fields) with
+        the raw output so expressions like ``{{score}} > 80`` can resolve, and
+        exposes the raw text as ``previous_output``.
+        """
+        context: Dict[str, Any] = {}
+        result = getattr(current_task, "result", None)
+        if result is not None:
+            try:
+                context.update(result.to_dict())
+            except Exception:
+                pass
+            context["previous_output"] = getattr(result, "raw", "")
+        return context
+
+    def _has_when_routing(self, task: Task) -> bool:
+        """True if the task uses the unified when/then_task/else_task routing."""
+        return task.when is not None or task.then_task is not None or task.else_task is not None
+
+    def _resolve_when_routing(self, current_task: Task) -> Optional[Task]:
+        """Resolve the next Task via the unified `when`/`then_task`/`else_task`
+        routing on the current task. Returns None if no such routing applies."""
+        if current_task.when is None and current_task.then_task is None and current_task.else_task is None:
+            return None
+        next_name = current_task.get_next_task(self._build_routing_context(current_task))
+        if not next_name:
+            return None
+        next_task = next((t for t in self.tasks.values() if t.name == next_name), None)
+        if next_task:
+            next_task.status = "not started"
+            logging.debug(f"Routing to {next_task.name} based on when-condition: {current_task.when}")
+        else:
+            logging.warning(f"when-routing target '{next_name}' not found among tasks")
+        return next_task
+
     def _build_task_context(self, current_task: Task) -> str:
         """Build context for a task based on its retain_full_context setting"""
         # Check if we have validation feedback to include
@@ -377,7 +418,7 @@ class Process:
         if current_task.retain_full_context:
             # Original behavior: include all previous tasks
             for prev_name in current_task.previous_tasks:
-                prev_task = next((t for t in self.tasks.values() if t.name == prev_name), None)
+                prev_task = self._task_by_name(prev_name)
                 if prev_task and prev_task.result:
                     context += f"\n{prev_name}: {prev_task.result.raw}"
                     
@@ -391,7 +432,7 @@ class Process:
             if current_task.previous_tasks:
                 # Get the most recent previous task (last in the list)
                 prev_name = current_task.previous_tasks[-1]
-                prev_task = next((t for t in self.tasks.values() if t.name == prev_name), None)
+                prev_task = self._task_by_name(prev_name)
                 if prev_task and prev_task.result:
                     context += f"\n{prev_name}: {prev_task.result.raw}"
                     
@@ -599,7 +640,7 @@ class Process:
         for task in self.tasks.values():
             if task.next_tasks:
                 for next_task_name in task.next_tasks:
-                    next_task = next((t for t in self.tasks.values() if t.name == next_task_name), None)
+                    next_task = self._task_by_name(next_task_name)
                     if next_task:
                         next_task.previous_tasks.append(task.name)
                         logging.debug(f"Added {task.name} as previous task for {next_task_name}")
@@ -770,7 +811,7 @@ Subtask: {st.name}
                             
                             target_tasks = current_task.condition.get(decision_str, []) if decision_str else []
                             task_value = target_tasks[0] if isinstance(target_tasks, list) else target_tasks
-                            next_task = next((t for t in self.tasks.values() if t.name == task_value), None)
+                            next_task = self._task_by_name(task_value)
                             if next_task:
                                 next_task.status = "not started"  # Reset status to allow execution
                                 logging.debug(f"Routing to {next_task.name} based on decision: {decision_str}")
@@ -799,7 +840,7 @@ Subtask: {st.name}
                         logging.debug(f"No input file, marking {current_task.name} as completed")
                         if current_task.next_tasks:
                             next_task_name = current_task.next_tasks[0]
-                            next_task = next((t for t in self.tasks.values() if t.name == next_task_name), None)
+                            next_task = self._task_by_name(next_task_name)
                             current_task = next_task
                         else:
                             current_task = None
@@ -811,8 +852,8 @@ Subtask: {st.name}
                 yield task_id
                 visited_tasks.add(task_id)
 
-                # Only end workflow if no next_tasks AND no conditions
-                if not current_task.next_tasks and not current_task.condition and not any(
+                # Only end workflow if no next_tasks AND no conditions AND no when-routing
+                if not current_task.next_tasks and not current_task.condition and not self._has_when_routing(current_task) and not any(
                     t.task_type == "loop" and current_task.name.startswith(t.name + "_")
                     for t in self.tasks.values()
                 ):
@@ -870,7 +911,7 @@ Subtask: {st.name}
                         else:
                             # Find the target task by name
                             task_value = target_tasks[0] if isinstance(target_tasks, list) else target_tasks
-                            next_task = next((t for t in self.tasks.values() if t.name == task_value), None)
+                            next_task = self._task_by_name(task_value)
                             if next_task:
                                 next_task.status = "not started"  # Reset status to allow execution
                                 
@@ -899,10 +940,26 @@ Subtask: {st.name}
                                 # Don't mark workflow as finished when following condition path
                                 await self._set_workflow_finished(False)
 
+            # Unified condition routing (when/then_task/else_task) takes priority
+            # over next_tasks when configured on the current task.
+            when_routing_exhausted = False
+            if not next_task and current_task:
+                if self._has_when_routing(current_task):
+                    when_task = self._resolve_when_routing(current_task)
+                    if when_task:
+                        next_task = when_task
+                        await self._set_workflow_finished(False)
+                    elif not current_task.next_tasks:
+                        # when-routing is authoritative: it resolved to no target
+                        # (e.g. the taken branch has no then/else_task) and there
+                        # is no next_tasks fallback, so this path ends cleanly
+                        # instead of picking an unrelated not-started task.
+                        when_routing_exhausted = True
+
             # If no condition-based routing, use next_tasks
             if not next_task and current_task and current_task.next_tasks:
                 next_task_name = current_task.next_tasks[0]
-                next_task = next((t for t in self.tasks.values() if t.name == next_task_name), None)
+                next_task = self._task_by_name(next_task_name)
                 if next_task:
                     # Reset the next task to allow re-execution
                     next_task.status = "not started"
@@ -915,7 +972,7 @@ Subtask: {st.name}
                     logging.debug(f"Following next_tasks to {next_task.name}")
 
             current_task = next_task
-            if not current_task:
+            if not current_task and not when_routing_exhausted:
                 current_task = self._find_next_not_started_task() # General fallback if no next task in workflow
 
 
@@ -956,7 +1013,19 @@ Workflow Finished: {self.workflow_finished} # ADDED: Workflow Finished Status
     async def asequential(self) -> AsyncGenerator[str, None]:
         """Async version of sequential method"""
         for task_id in self.tasks:
-            if self.tasks[task_id].status != "completed":
+            task = self.tasks[task_id]
+            if task.status == "failed":
+                continue  # Skip failed tasks
+            # Check if any context dependency failed
+            if hasattr(task, 'context') and task.context:
+                deps_failed = any(
+                    self.tasks[dep.id].status == "failed"
+                    for dep in task.context if hasattr(dep, 'id') and dep.id in self.tasks
+                )
+                if deps_failed:
+                    task.status = "failed"
+                    continue
+            if task.status != "completed":
                 yield task_id
 
     async def ahierarchical(self) -> AsyncGenerator[str, None]:
@@ -1154,7 +1223,7 @@ Provide a JSON with the structure:
         for task in self.tasks.values():
             if task.next_tasks:
                 for next_task_name in task.next_tasks:
-                    next_task = next((t for t in self.tasks.values() if t.name == next_task_name), None)
+                    next_task = self._task_by_name(next_task_name)
                     if next_task:
                         next_task.previous_tasks.append(task.name)
 
@@ -1349,7 +1418,7 @@ Subtask: {st.name}
                             
                             target_tasks = current_task.condition.get(decision_str, []) if decision_str else []
                             task_value = target_tasks[0] if isinstance(target_tasks, list) else target_tasks
-                            next_task = next((t for t in self.tasks.values() if t.name == task_value), None)
+                            next_task = self._task_by_name(task_value)
                             if next_task:
                                 next_task.status = "not started"  # Reset status to allow execution
                                 logging.debug(f"Routing to {next_task.name} based on decision: {decision_str}")
@@ -1378,7 +1447,7 @@ Subtask: {st.name}
                         logging.debug(f"No input file, marking {current_task.name} as completed")
                         if current_task.next_tasks:
                             next_task_name = current_task.next_tasks[0]
-                            next_task = next((t for t in self.tasks.values() if t.name == next_task_name), None)
+                            next_task = self._task_by_name(next_task_name)
                             current_task = next_task
                         else:
                             current_task = None
@@ -1390,8 +1459,8 @@ Subtask: {st.name}
                 yield task_id
                 visited_tasks.add(task_id)
 
-                # Only end workflow if no next_tasks AND no conditions
-                if not current_task.next_tasks and not current_task.condition and not any(
+                # Only end workflow if no next_tasks AND no conditions AND no when-routing
+                if not current_task.next_tasks and not current_task.condition and not self._has_when_routing(current_task) and not any(
                     t.task_type == "loop" and current_task.name.startswith(t.name + "_")
                     for t in self.tasks.values()
                 ):
@@ -1447,7 +1516,7 @@ Subtask: {st.name}
                         else:
                             # Find the target task by name
                             task_value = target_tasks[0] if isinstance(target_tasks, list) else target_tasks
-                            next_task = next((t for t in self.tasks.values() if t.name == task_value), None)
+                            next_task = self._task_by_name(task_value)
                             if next_task:
                                 next_task.status = "not started"  # Reset status to allow execution
                                 
@@ -1476,10 +1545,26 @@ Subtask: {st.name}
                                 # Don't mark workflow as finished when following condition path
                                 self._set_workflow_finished_sync(False)
 
+            # Unified condition routing (when/then_task/else_task) takes priority
+            # over next_tasks when configured on the current task.
+            when_routing_exhausted = False
+            if not next_task and current_task:
+                if self._has_when_routing(current_task):
+                    when_task = self._resolve_when_routing(current_task)
+                    if when_task:
+                        next_task = when_task
+                        self._set_workflow_finished_sync(False)
+                    elif not current_task.next_tasks:
+                        # when-routing is authoritative: it resolved to no target
+                        # (e.g. the taken branch has no then/else_task) and there
+                        # is no next_tasks fallback, so this path ends cleanly
+                        # instead of picking an unrelated not-started task.
+                        when_routing_exhausted = True
+
             # If no condition-based routing, use next_tasks
             if not next_task and current_task and current_task.next_tasks:
                 next_task_name = current_task.next_tasks[0]
-                next_task = next((t for t in self.tasks.values() if t.name == next_task_name), None)
+                next_task = self._task_by_name(next_task_name)
                 if next_task:
                     # Reset the next task to allow re-execution
                     next_task.status = "not started"
@@ -1492,7 +1577,7 @@ Subtask: {st.name}
                     logging.debug(f"Following next_tasks to {next_task.name}")
 
             current_task = next_task
-            if not current_task:
+            if not current_task and not when_routing_exhausted:
                 current_task = self._find_next_not_started_task() # General fallback if no next task in workflow