diff --git a/main.py b/main.py index cd8a45b..715e38f 100644 --- a/main.py +++ b/main.py @@ -1,124 +1,142 @@ import os import asyncio +import json from pathlib import Path -from langchain_openai import ChatOpenAI, OpenAIEmbeddings -from langchain_chroma import Chroma -from langchain_core.documents import Document +from typing import List + +from langchain_openai import ChatOpenAI +from langchain_core.messages import HumanMessage from langchain.tools import tool from deepagents import create_deep_agent from deepagents.backends import FilesystemBackend, LocalShellBackend, CompositeBackend -# ----------------- Configuration ----------------- -# Load API key from .env or environment variable -os.environ.setdefault("OPENAI_API_KEY", os.getenv("OPENAI_API_KEY", "")) +# --- Embeddings and vector store (Qdrant + Ollama) --- +from langchain_ollama import OllamaEmbeddings +from langchain_qdrant import Qdrant +from langchain_core.documents import Document -# LLM via OpenRouter +# Load OpenRouter key for LLM (required by Ollama embeddings as well) +OPENAI_API_KEY = os.getenv("OPENAI_API_KEY") + +# LLM configuration (OpenRouter) llm = ChatOpenAI( model="openai/gpt-oss-20b:free", base_url="https://openrouter.ai/api/v1", - api_key=os.getenv("OPENAI_API_KEY"), + api_key=OPENAI_API_KEY, temperature=0.0, ) -# Embeddings for Chroma -embeddings = OpenAIEmbeddings( - model="text-embedding-3-small", - base_url="https://openrouter.ai/api/v1", - api_key=os.getenv("OPENAI_API_KEY"), -) +# Embeddings via Ollama (nomic-embed-text) +embeddings = OllamaEmbeddings(model="nomic-embed-text") -# ----------------- Chroma DB ----------------- -CHROMA_PATH = Path("./chroma_faq") -CHROMA_COLLECTION = "faq_collection" -vector_store = Chroma( - collection_name=CHROMA_COLLECTION, +# Qdrant client (in‑memory for demo; replace with host/port for prod) +qdrant_client = Qdrant( + collection_name="faq_collection", + url="http://localhost:6333", # default Qdrant local URL embedding_function=embeddings, - persist_directory=str(CHROMA_PATH), ) -# Load markdown files into Chroma (idempotent) -DATA_DIR = Path("./data") -if not CHROMA_PATH.exists() or not list(CHROMA_PATH.iterdir()): - docs = [] +# --- Data loading and indexing --- +DATA_DIR = Path("data") +CHROMA_PERSIST = Path("./qdrant_faq") # not used directly but kept for compatibility + + +def load_faq_to_qdrant() -> None: + """Read all .md files from DATA_DIR, chunk them, embed and store in Qdrant.""" + if not DATA_DIR.exists(): + print("Data directory not found. Create 'data/' with .md files.") + return + docs: List[Document] = [] for md_file in DATA_DIR.glob("*.md"): text = md_file.read_text(encoding="utf-8") - docs.append(Document(page_content=text, metadata={"source": md_file.name})) - vector_store.add_documents(docs) - vector_store.persist() + # Simple split by double newlines as a naive chunker + for i, chunk in enumerate(text.split("\n\n")): + docs.append(Document(page_content=chunk, metadata={"source": md_file.name, "chunk": i})) + # Add to Qdrant + qdrant_client.add_documents(docs) + print(f"Indexed {len(docs)} chunks into Qdrant.") -# ----------------- Tools ----------------- +# --- Tools --- @tool -def search_course_docs(query: str) -> str: +def search_course_docs(query: str, k: int = 3) -> str: """Search the local FAQ collection for relevant passages.""" - results = vector_store.similarity_search(query, k=3) + results = qdrant_client.similarity_search(query, k=k) if not results: return "No relevant information found in the course materials." - return "\n---\n".join(f"{doc.metadata.get('source', 'unknown')}\n{doc.page_content}" for doc in results) + return "\n\n---\n\n".join(r.page_content for r in results) -# Mock MCP tool – static JSON data -COURSE_META = { - "schedule": "Monday 10:00-12:00, Wednesday 14:00-16:00", - "instructor": "Dr. Ivanov", - "location": "Room 101", -} +# MCP-style tool: fetch metadata from a static JSON file +META_FILE = Path("meta.json") @tool def fetch_course_meta(query: str) -> str: - """Return course metadata matching the query keyword. - For example, query="schedule" returns the schedule string. + """Return course metadata that matches the query. + For demo purposes, we load a static JSON file and perform a simple keyword search. """ - key = query.lower().strip() - return COURSE_META.get(key, f"No metadata found for '{query}'.") + if not META_FILE.exists(): + return "Metadata file not found." + data = json.loads(META_FILE.read_text(encoding="utf-8")) + # Very naive matching: return entries where query is a substring of any value + matches = [] + for key, value in data.items(): + if isinstance(value, str) and query.lower() in value.lower(): + matches.append(f"{key}: {value}") + if not matches: + return "No metadata matches your query." + return "\n".join(matches) -# ----------------- Backend ----------------- +# --- Agent setup --- backend = CompositeBackend([ LocalShellBackend(workspace_dir="./workspace"), FilesystemBackend(), ]) -# ----------------- Agent ----------------- agent = create_deep_agent( model=llm, tools=[search_course_docs, fetch_course_meta], backend=backend, system_prompt=( - "You are a helpful FAQ assistant for the course.\n" - "When answering a question, first decide whether the answer comes from the course materials (use search_course_docs)\n" - "or from course metadata (use fetch_course_meta).\n" + "You are a helpful FAQ bot for a course.\n" + "When a user asks about course materials, use the search_course_docs tool.\n" + "When a user asks about schedule, metadata, or other non‑material info, use fetch_course_meta.\n" "Do not call both tools unless absolutely necessary.\n" - "In your final answer, prepend 'source: chroma' or 'source: mcp_meta' to indicate the origin." + "In your final answer, prepend 'source: chroma' if you used search_course_docs,\n" + "or 'source: mcp_meta' if you used fetch_course_meta." ), ) -# ----------------- CLI ----------------- +# --- CLI --- PRESET_QUESTIONS = [ - "What topics are covered in the first lecture?", # should hit chroma - "Who is the instructor for this course?", # should hit mcp_meta - "When is the next class?", # should hit chroma (or meta if schedule) + "What topics are covered in the first lecture?", + "How can I access the lecture slides?", + "What is the schedule for the next week?", ] -async def run_question(question: str): +async def run_agent(question: str, thread_id: str = "session-1"): result = await agent.ainvoke( - {"messages": ["HumanMessage(content=\"{}\")".format(question)]}, - {"configurable": {"thread_id": "session-1"}}, + {"messages": [HumanMessage(content=question)]}, + {"configurable": {"thread_id": thread_id}}, ) - # The agent returns a dict with 'messages'; extract last content - content = result["messages"][-1].content - print(f"\nQ: {question}\nA: {content}\n") - -async def interactive(): - print("Enter a question (or 'exit' to quit):") - while True: - q = input("> ") - if q.lower() in {"exit", "quit"}: - break - await run_question(q) + return result["messages"][-1].content async def main(): - print("Running preset questions...") - for q in PRESET_QUESTIONS: - await run_question(q) - await interactive() + # Ensure the vector store is populated + load_faq_to_qdrant() + + print("\n--- FAQ Bot Demo ---\n") + for i, q in enumerate(PRESET_QUESTIONS, 1): + print(f"Q{i}: {q}") + answer = await run_agent(q, thread_id=f"demo-{i}") + print(f"A{i}: {answer}\n") + + # Interactive mode + print("Enter your own questions (type 'exit' to quit):") + while True: + user_q = input("> ") + if user_q.lower() in {"exit", "quit"}: + break + answer = await run_agent(user_q, thread_id="interactive") + print(answer) if __name__ == "__main__": asyncio.run(main())