update

agent.py

@@ -41,16 +41,21 @@ if backend_parent not in sys.path:
 
 # Import backend modules (support both module and script execution)
 try:
-    from .config import get_config
-    from .retrieve import search as retrieve_search
-    from .logging_config import setup_logging
+    from config import get_config
+    from retrieve import search as retrieve_search
+    from logging_config import setup_logging
 except ImportError as e:
     try:
-        from backend.config import get_config
-        from backend.retrieve import search as retrieve_search
-        from backend.logging_config import setup_logging
+        from .config import get_config
+        from .retrieve import search as retrieve_search
+        from .logging_config import setup_logging
     except ImportError as e2:
-        raise ImportError(f"Failed to import backend modules: {e2}")
+        try:
+            from backend.config import get_config
+            from backend.retrieve import search as retrieve_search
+            from backend.logging_config import setup_logging
+        except ImportError as e3:
+            raise ImportError(f"Failed to import backend modules: {e3}")
 
 # Import OpenAI Agents SDK (must be installed separately)
 try:

api.py

@@ -38,7 +38,7 @@ except ImportError as e:
 app = FastAPI(
     title="RAG Chatbot API",
     version="1.0.0",
-    description="FastAPI wrapper for RAG Book Assistant"
+    description="FastAPI wrapper for RAG Book Assistant",
 )
 
 # ============ CORS Configuration ============
@@ -62,7 +62,7 @@ app.add_middleware(
 class ChatRequest(BaseModel):
     question: str = Field(..., min_length=1, max_length=1000)
 
-    @validator('question')
+    @validator("question")
     def validate_question(cls, v):
         if not v or not v.strip():
             raise ValueError("Question cannot be empty")
@@ -91,10 +91,12 @@ class HealthStatus(BaseModel):
 
 # ============ Health Check ============
 
+
 def check_qdrant_health() -> str:
     try:
-        from backend.config import get_config
+        from config import get_config
         from qdrant_client import QdrantClient
+
         cfg = get_config()
         client = QdrantClient(url=cfg["qdrant_url"], api_key=cfg["qdrant_api_key"])
         client.get_collection(cfg["qdrant_collection"])
@@ -109,6 +111,7 @@ def check_openai_health() -> str:
         if not api_key:
             return "disconnected"
         import openai
+
         client = openai.OpenAI(api_key=api_key)
         client.models.list()
         return "connected"
@@ -120,17 +123,20 @@ def check_openai_health() -> str:
 async def health_check():
     qdrant = check_qdrant_health()
     openai = check_openai_health()
-    status = "healthy" if qdrant == "connected" and openai == "connected" else "degraded"
+    status = (
+        "healthy" if qdrant == "connected" and openai == "connected" else "degraded"
+    )
     return HealthStatus(
         status=status,
         qdrant=qdrant,
         openai=openai,
-        timestamp=datetime.utcnow().isoformat() + "Z"
+        timestamp=datetime.utcnow().isoformat() + "Z",
     )
 
 
 # ============ Chat Endpoint ============
 
+
 @app.post("/chat")
 async def chat_endpoint(request: ChatRequest):
     request_id = str(uuid.uuid4())[:8]
@@ -140,10 +146,7 @@ async def chat_endpoint(request: ChatRequest):
         agent = get_agent()
 
         # Run agent with timeout (20s to accommodate full workflow)
-        result = await asyncio.wait_for(
-            Runner.run(agent, question),
-            timeout=20.0
-        )
+        result = await asyncio.wait_for(Runner.run(agent, question), timeout=20.0)
 
         # Extract sources from tool call outputs
         sources = []
@@ -153,42 +156,54 @@ async def chat_endpoint(request: ChatRequest):
                     output = item.output
                     if isinstance(output, list):
                         for chunk in output:
-                            sources.append(Source(
-                                url=chunk.get("url", ""),
-                                chunk_index=chunk.get("chunk_index", 0),
-                                text_snippet=chunk.get("text", "")[:200]
-                            ))
+                            sources.append(
+                                Source(
+                                    url=chunk.get("url", ""),
+                                    chunk_index=chunk.get("chunk_index", 0),
+                                    text_snippet=chunk.get("text", "")[:200],
+                                )
+                            )
 
         # Get token usage
         tokens_used = 0
-        if result.context_wrapper and hasattr(result.context_wrapper, 'usage'):
+        if result.context_wrapper and hasattr(result.context_wrapper, "usage"):
             tokens_used = result.context_wrapper.usage.total_tokens
 
         return ChatResponse(
             answer=result.final_output,
             sources=sources,
             tokens_used=tokens_used,
-            agent_trace=f"{request_id}: completed"
+            agent_trace=f"{request_id}: completed",
         )
 
     except asyncio.TimeoutError:
         return JSONResponse(
             status_code=504,
-            content={"error": "timeout", "message": "The chatbot is taking too long to respond. Please try a shorter question."}
+            content={
+                "error": "timeout",
+                "message": "The chatbot is taking too long to respond. Please try a shorter question.",
+            },
         )
 
     except Exception as e:
         if "openai" in str(e).lower() or "rate limit" in str(e).lower():
             return JSONResponse(
                 status_code=503,
-                content={"error": "openai_failed", "message": "The AI service is currently unavailable. Please try again in a few minutes."}
+                content={
+                    "error": "openai_failed",
+                    "message": "The AI service is currently unavailable. Please try again in a few minutes.",
+                },
             )
         return JSONResponse(
             status_code=500,
-            content={"error": "internal_error", "message": "An unexpected error occurred. Please refresh the page and try again."}
+            content={
+                "error": "internal_error",
+                "message": "An unexpected error occurred. Please refresh the page and try again.",
+            },
         )
 
 
 if __name__ == "__main__":
     import uvicorn
+
     uvicorn.run(app, host="0.0.0.0", port=8000)