Initial commit: Multi-service AI agent system

- Frontend: Vite + React + TypeScript chat interface
- Backend: FastAPI gateway with LangGraph routing
- Knowledge Service: ChromaDB RAG with Gitea scraper
- LangGraph Service: Multi-agent orchestration
- Airflow: Scheduled Gitea ingestion DAG
- Documentation: Complete plan and implementation guides

Architecture:
- Modular Docker Compose per service
- External ai-mesh network for communication
- Fast rebuilds with /app/packages pattern
- Intelligent agent routing (no hardcoded keywords)

Services:
- Frontend (5173): React chat UI
- Chat Gateway (8000): FastAPI entry point
- LangGraph (8090): Agent orchestration
- Knowledge (8080): ChromaDB RAG
- Airflow (8081): Scheduled ingestion
- PostgreSQL (5432): Chat history

Excludes: node_modules, .venv, chroma_db, logs, .env files
Includes: All source code, configs, docs, docker files

backend/main.py

@@ -0,0 +1,58 @@
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+import httpx
+import logging
+import sys
+import traceback
+
+logging.basicConfig(level=logging.INFO, format="%(asctime)s [%(levelname)s] %(message)s", handlers=[logging.StreamHandler(sys.stdout)])
+logger = logging.getLogger(__name__)
+
+app = FastAPI()
+app.add_middleware(CORSMiddleware, allow_origins=["*"], allow_credentials=True, allow_methods=["*"], allow_headers=["*"])
+
+class MessageRequest(BaseModel):
+    message: str
+
+BRAIN_URL = "http://opencode-brain:5000"
+KNOWLEDGE_URL = "http://knowledge-service:8080/query"
+AUTH = httpx.BasicAuth("opencode", "sam4jo")
+
+@app.post("/chat")
+async def chat(request: MessageRequest):
+    user_msg = request.message.lower()
+    timeout_long = httpx.Timeout(180.0, connect=10.0)
+    timeout_short = httpx.Timeout(5.0, connect=2.0)
+    
+    context = ""
+    # Check for keywords to trigger Librarian (DB) lookup
+    if any(kw in user_msg for kw in ["sam", "hobby", "music", "guitar", "skiing", "experience"]):
+        logger.info("Gateway: Consulting Librarian (DB)...")
+        async with httpx.AsyncClient(timeout=timeout_short) as client:
+            try:
+                k_res = await client.post(KNOWLEDGE_URL, json={"question": request.message})
+                if k_res.status_code == 200:
+                    context = k_res.json().get("context", "")
+            except Exception as e:
+                logger.warning(f"Gateway: Librarian offline/slow: {str(e)}")
+
+    # Forward to Brain (LLM)
+    async with httpx.AsyncClient(auth=AUTH, timeout=timeout_long) as brain_client:
+        try:
+            session_res = await brain_client.post(f"{BRAIN_URL}/session", json={"title": "Demo"})
+            session_id = session_res.json()["id"]
+            final_prompt = f"CONTEXT:\n{context}\n\nUSER: {request.message}" if context else request.message
+            response = await brain_client.post(f"{BRAIN_URL}/session/{session_id}/message", json={"parts": [{"type": "text", "text": final_prompt}]})
+            
+            # FIX: Iterate through parts array to find text response
+            data = response.json()
+            if "parts" in data:
+                for part in data["parts"]:
+                    if part.get("type") == "text" and "text" in part:
+                        return {"response": part["text"]}
+            
+            return {"response": "AI responded but no text found in expected format."}
+        except Exception:
+            logger.error(f"Gateway: Brain failure: {traceback.format_exc()}")
+            return {"response": "Error: The Brain is taking too long or is disconnected."}