Log task planner conversations locally

apps/device-host-agent/host_agent/app.py

@@ -231,12 +231,14 @@ def create_application(
                 "MCP token generated at %s", mcp_token_path
             )
         mcp_busy_tracker = McpBusyTracker(ttl_seconds=20.0)
+        conversation_log = ConversationLogStore(resolved_config.conversation_log_path) if resolved_config.mode == "local" else None
         executor = AssignmentExecutor(
             create_execution_factories(
                 resolved_manager,
                 metadata_store=metadata_store,
                 timeline=timeline,
                 host_agent_config=resolved_config,
+                conversation_log=conversation_log,
             ),
             mcp_busy_tracker=mcp_busy_tracker,
         )
@@ -246,7 +248,6 @@ def create_application(
             status_tracker=status_tracker,
         )
         planner_config = load_planner_config()
-        conversation_log = ConversationLogStore(resolved_config.conversation_log_path) if resolved_config.mode == "local" else None
         conversation_agent = (
             ConversationAgent(
                 config=planner_config,

apps/device-host-agent/host_agent/execution.py

@@ -3,6 +3,7 @@ from __future__ import annotations
 import os
 from collections.abc import Callable
 from dataclasses import dataclass, replace
+from typing import Any
 
 from device.manager import DeviceManager
 from host_agent.cloud_planner_client import CloudProxyToolCallingClient
@@ -35,6 +36,7 @@ def create_execution_factories(
     metadata_store: TaskMetadataStore | None = None,
     timeline: Timeline | None = None,
     host_agent_config: HostAgentConfig | None = None,
+    conversation_log: Any | None = None,
 ) -> ExecutionFactories:
     shared_workflow_store = workflow_store or WorkflowStore()
     resolved_host_agent_config = host_agent_config
@@ -48,7 +50,10 @@ def create_execution_factories(
             ),
             metadata_store=metadata_store,
             timeline=timeline,
-            planner=_host_agent_planner(resolved_host_agent_config),
+            planner=_host_agent_planner(
+                resolved_host_agent_config,
+                event_logger=conversation_log.append if conversation_log is not None else None,
+            ),
             planner_config=_host_agent_planner_config(),
             device_platform_provider=lambda device_id: _device_platform(
                 manager, device_id
@@ -86,6 +91,8 @@ def _host_agent_planner_config() -> PlannerConfig:
 
 def _host_agent_planner(
     host_agent_config: HostAgentConfig | None,
+    *,
+    event_logger: Callable[[dict[str, Any]], None] | None = None,
 ) -> Planner | None:
     """Build the `AIPlanner` explicitly when the cloud-proxy transport is
     selected, so its `ToolCallingClient` is a `CloudProxyToolCallingClient`
@@ -100,11 +107,12 @@ def _host_agent_planner(
 
     resolved_config = host_agent_config or load_host_agent_config()
     if resolved_config.ai_planner_transport != "cloud":
-        return None
+        return AIPlanner(config=planner_config, event_logger=event_logger)
 
     return AIPlanner(
         client=CloudProxyToolCallingClient(resolved_config),
         config=planner_config,
+        event_logger=event_logger,
     )
 
 

runtime/ai_planner.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from collections.abc import Callable
 from typing import TYPE_CHECKING, Any
 
 from core.errors import TaskFailedError
@@ -23,9 +24,11 @@ class AIPlanner(Planner):
         *,
         client: ToolCallingClient | None = None,
         config: PlannerConfig | None = None,
+        event_logger: Callable[[dict[str, Any]], None] | None = None,
     ) -> None:
         self.config = config or load_config()
         self.client = client or build_client(self.config)
+        self.event_logger = event_logger
 
     def plan(
         self,
@@ -49,13 +52,24 @@ class AIPlanner(Planner):
                 "corrected action or finish the task if it cannot proceed.\n"
                 f"Previous failure: {context.step_results[-1].error or 'unknown error'}"
             )
-        decision = self.client.decide(
-            system_prompt=PLANNER_SYSTEM_PROMPT,
-            user_prompt=user_prompt,
-            screenshot=screenshot,
-            tools=ALL_TOOL_SPECS,
-            timeout=self.config.timeout,
-        )
+        self._log({"type": "llm_request", "task_id": context.task_id, "goal": goal,
+                   "system_prompt": PLANNER_SYSTEM_PROMPT, "user_prompt": user_prompt,
+                   "has_screenshot": screenshot is not None})
+        try:
+            decision = self.client.decide(
+                system_prompt=PLANNER_SYSTEM_PROMPT,
+                user_prompt=user_prompt,
+                screenshot=screenshot,
+                tools=ALL_TOOL_SPECS,
+                timeout=self.config.timeout,
+            )
+        except Exception as exc:
+            self._log({"type": "agent_error", "task_id": context.task_id, "error": str(exc)})
+            raise
+        self._log({"type": "llm_response", "task_id": context.task_id,
+                   "content": decision.text_output, "thinking": decision.thinking,
+                   "tool_name": decision.tool_name, "arguments": decision.arguments,
+                   "purpose": decision.purpose, "expected_outcome": decision.expected_outcome})
 
         if decision.tool_name == FINISH_TASK_TOOL:
             if decision.arguments.get("success"):
@@ -85,6 +99,13 @@ class AIPlanner(Planner):
         # (mapped to an empty plan above), never via this hook.
         return False
 
+    def _log(self, event: dict[str, Any]) -> None:
+        if self.event_logger is not None:
+            try:
+                self.event_logger(event)
+            except Exception:
+                pass
+
 
 def _history_summary(world: "WorldState | None") -> list[dict[str, Any]]:
     if world is None: