Fix Cain 'unknown' error: Add comprehensive startup initialization and error logging

- Add ensure_cain_status_file() to create/validate status file on startup
- Fix invalid 'unknown' state by resetting to 'idle'
- Add startup error/warning logging with get_startup_report()
- Add /api/agent/startup endpoint for diagnostics
- Add /api/agent/reset-status endpoint for manual status reset
- Print startup summary on initialization

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

app.py

@@ -46,6 +46,131 @@ sys.path.insert(0, str(BASE_DIR))
 # File paths
 CAIN_STATUS_FILE = AGENTS_DIR / "cain_status.json"
 REGISTRY_FILE = AGENTS_DIR / "registry.json"
+LOGS_DIR = AGENTS_DIR / "logs"
+
+# ========== Startup Error Logger ==========
+
+_startup_errors = []
+_startup_warnings = []
+
+def log_startup_error(component: str, error: str, details: Dict[str, Any] = None):
+    """Log a startup error for later reporting."""
+    _startup_errors.append({
+        "component": component,
+        "error": error,
+        "details": details or {},
+        "timestamp": datetime.utcnow().isoformat() + "Z"
+    })
+    print(f"[STARTUP ERROR] {component}: {error}", file=sys.stderr)
+    if details:
+        print(f"  Details: {details}", file=sys.stderr)
+
+def log_startup_warning(component: str, warning: str, details: Dict[str, Any] = None):
+    """Log a startup warning for later reporting."""
+    _startup_warnings.append({
+        "component": component,
+        "warning": warning,
+        "details": details or {},
+        "timestamp": datetime.utcnow().isoformat() + "Z"
+    })
+    print(f"[STARTUP WARNING] {component}: {warning}")
+
+def get_startup_report() -> Dict[str, Any]:
+    """Get a comprehensive startup report."""
+    return {
+        "timestamp": datetime.utcnow().isoformat() + "Z",
+        "base_dir": str(BASE_DIR),
+        "agents_dir": str(AGENTS_DIR),
+        "cain_status_file": str(CAIN_STATUS_FILE),
+        "errors": _startup_errors,
+        "warnings": _startup_warnings,
+        "error_count": len(_startup_errors),
+        "warning_count": len(_startup_warnings)
+    }
+
+# ========== Status File Initialization ==========
+
+def ensure_cain_status_file() -> Dict[str, Any]:
+    """
+    Ensure cain_status.json exists and is properly initialized.
+    Creates the file if it doesn't exist with a valid default status.
+    Returns the current status.
+    """
+    AGENTS_DIR.mkdir(parents=True, exist_ok=True)
+    LOGS_DIR.mkdir(parents=True, exist_ok=True)
+
+    default_status = {
+        "current_state": "idle",
+        "last_updated": datetime.utcnow().isoformat() + "Z",
+        "agent": "cain",
+        "startup_time": datetime.utcnow().isoformat() + "Z"
+    }
+
+    if not CAIN_STATUS_FILE.exists():
+        try:
+            with open(CAIN_STATUS_FILE, 'w') as f:
+                json.dump(default_status, f, indent=2)
+            log_startup_warning("cain_status", "Created default cain_status.json file", {
+                "path": str(CAIN_STATUS_FILE),
+                "status": default_status
+            })
+        except Exception as e:
+            log_startup_error("cain_status", f"Failed to create status file: {e}", {
+                "path": str(CAIN_STATUS_FILE),
+                "error_type": type(e).__name__
+            })
+        return default_status
+
+    # File exists - validate and fix if needed
+    try:
+        with open(CAIN_STATUS_FILE, 'r') as f:
+            status = json.load(f)
+
+        # Ensure required fields exist
+        updated = False
+        if "current_state" not in status:
+            status["current_state"] = "idle"
+            updated = True
+        if "last_updated" not in status:
+            status["last_updated"] = datetime.utcnow().isoformat() + "Z"
+            updated = True
+        if "agent" not in status:
+            status["agent"] = "cain"
+            updated = True
+
+        # Fix invalid "unknown" state
+        if status.get("current_state") == "unknown":
+            old_state = status["current_state"]
+            status["current_state"] = "idle"
+            updated = True
+            log_startup_warning("cain_status", f"Fixed invalid state '{old_state}' -> 'idle'", {
+                "old_state": old_state,
+                "new_state": "idle"
+            })
+
+        if updated:
+            with open(CAIN_STATUS_FILE, 'w') as f:
+                json.dump(status, f, indent=2)
+            log_startup_warning("cain_status", "Updated cain_status.json with missing fields")
+
+        return status
+
+    except json.JSONDecodeError as e:
+        log_startup_error("cain_status", f"Invalid JSON in status file, recreating: {e}", {
+            "path": str(CAIN_STATUS_FILE)
+        })
+        try:
+            with open(CAIN_STATUS_FILE, 'w') as f:
+                json.dump(default_status, f, indent=2)
+        except Exception:
+            pass
+        return default_status
+    except Exception as e:
+        log_startup_error("cain_status", f"Error reading status file: {e}", {
+            "path": str(CAIN_STATUS_FILE),
+            "error_type": type(e).__name__
+        })
+        return default_status
 
 # ========== WebSocket Manager ==========
 
@@ -896,6 +1021,11 @@ def create_agent_office_with_ws():
     Create the Agent Office with WebSocket API endpoints.
     Returns a FastAPI app with the Gradio app mounted at root.
     """
+    # ========== Startup Initialization ==========
+    print("[Agent Office] Initializing startup...")
+    initial_status = ensure_cain_status_file()
+    print(f"[Agent Office] Initial status: {initial_status.get('current_state', 'unknown')}")
+
     # Create FastAPI app first
     fastapi_app = FastAPI(title="HuggingClaw Agent Office API")
 
@@ -926,6 +1056,20 @@ def create_agent_office_with_ws():
         """Get agent registry as JSON."""
         return load_agent_registry()
 
+    @fastapi_app.get("/api/agent/startup")
+    async def api_agent_startup():
+        """Get startup initialization report."""
+        return get_startup_report()
+
+    @fastapi_app.post("/api/agent/reset-status")
+    async def api_reset_status():
+        """Reset cain status file to idle."""
+        try:
+            status = ensure_cain_status_file()
+            return {"success": True, "status": status}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
     # ========== Analytics API Endpoints ==========
     if ANALYTICS_AVAILABLE:
         @fastapi_app.get("/api/analytics/stats")
@@ -982,6 +1126,16 @@ if __name__ == "__main__":
     print(f"[Agent Office] Base directory: {BASE_DIR}")
     print(f"[Agent Office] WebSocket API enabled at /api/thoughts")
 
+    # Print startup report
+    startup_report = get_startup_report()
+    print(f"[Agent Office] Startup Report:")
+    print(f"  - Errors: {startup_report['error_count']}")
+    print(f"  - Warnings: {startup_report['warning_count']}")
+    if startup_report['error_count'] > 0:
+        print("[Agent Office] ERROR: Startup errors detected!")
+        for err in startup_report['errors']:
+            print(f"  - {err['component']}: {err['error']}")
+
     # Use uvicorn to serve the FastAPI app (not .launch())
     import uvicorn
     uvicorn.run(app, host=server_name, port=server_port)