Method _triage_query_async

rag_system/agent/loop.py:171–219

(self, query: str, history: list)

    # ---------------- Asynchronous triage using Ollama ----------------
    async def _triage_query_async(self, query: str, history: list) -> str:
        
        print(f"🔍 ROUTING DEBUG: Starting triage for query: '{query[:100]}...'")
        
        # 1️⃣ Fast routing using precomputed overviews (if available)
        print(f"📖 ROUTING DEBUG: Attempting overview-based routing...")
        routed = self._route_via_overviews(query)
        if routed:
            print(f"✅ ROUTING DEBUG: Overview routing decided: '{routed}'")
            return routed
        else:
            print(f"❌ ROUTING DEBUG: Overview routing returned None, falling back to LLM triage")

        if history:
            # If there's history, the query is likely a follow-up, so we default to RAG.
            # A more advanced implementation could use an LLM to see if the new query
            # changes the topic entirely.
            print(f"📜 ROUTING DEBUG: History exists, defaulting to 'rag_query'")
            return "rag_query"

        print(f"🤖 ROUTING DEBUG: No history, using LLM fallback triage...")
        prompt = f"""
You are a query routing expert. Analyze the user's question and decide which backend should handle it.

Choose **exactly one** category:

1. "rag_query" – Questions about the user's uploaded documents or specific document content that should be searched. Examples: "What is the invoice amount?", "Summarize the research paper", "What companies are mentioned?"

2. "direct_answer" – General knowledge questions, greetings, or queries unrelated to uploaded documents. Examples: "Who are the CEOs of Tesla and Amazon?", "What is the capital of France?", "Hello", "Explain quantum physics"

3. "graph_query" – Specific factual relations for knowledge-graph lookup (currently limited use)

IMPORTANT: For general world knowledge about well-known companies, people, or facts NOT related to uploaded documents, choose "direct_answer".

User query: "{query}"

Respond with JSON: {{"category": "<your_choice>"}}
"""
        resp = self.llm_client.generate_completion(
            model=self.ollama_config["generation_model"], prompt=prompt, format="json"
        )
        try:
            data = json.loads(resp.get("response", "{}"))
            decision = data.get("category", "rag_query")
            print(f"🤖 ROUTING DEBUG: LLM fallback triage decided: '{decision}'")
            return decision
        except json.JSONDecodeError:
            print(f"❌ ROUTING DEBUG: LLM fallback triage JSON parsing failed, defaulting to 'rag_query'")
            return "rag_query"

    def _run_graph_query(self, query: str, history: list) -> Dict[str, Any]:
        contextual_query = self._format_query_with_history(query, history)