139 lines
5.6 KiB
Python
139 lines
5.6 KiB
Python
import os
|
|
import asyncio
|
|
from pathlib import Path
|
|
|
|
from dotenv import load_dotenv
|
|
from langchain_openai import ChatOpenAI, OpenAIEmbeddings
|
|
from langchain_core.documents import Document
|
|
from langchain_text_splitters import RecursiveCharacterTextSplitter
|
|
from langchain_qdrant import QdrantVectorStore
|
|
from langchain.tools import tool
|
|
from deepagents import create_deep_agent
|
|
from deepagents.backends import CompositeBackend, LocalShellBackend, FilesystemBackend
|
|
|
|
# Load environment variables (OPENAI_API_KEY)
|
|
load_dotenv()
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# LLM and embeddings configuration (OpenRouter)
|
|
# ---------------------------------------------------------------------------
|
|
llm = ChatOpenAI(
|
|
model="openai/gpt-oss-20b:free",
|
|
base_url="https://openrouter.ai/api/v1",
|
|
api_key=os.getenv("OPENAI_API_KEY"),
|
|
temperature=0.0,
|
|
)
|
|
|
|
embeddings = OpenAIEmbeddings(
|
|
model="text-embedding-3-small",
|
|
base_url="https://openrouter.ai/api/v1",
|
|
api_key=os.getenv("OPENAI_API_KEY"),
|
|
)
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Qdrant vector store setup
|
|
# ---------------------------------------------------------------------------
|
|
from qdrant_client import QdrantClient
|
|
|
|
qdrant_client = QdrantClient(url="http://localhost:6333") # Qdrant must be running locally
|
|
vector_store = QdrantVectorStore(
|
|
embedding_function=embeddings,
|
|
client=qdrant_client,
|
|
collection_name="knowledge",
|
|
)
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Text splitter for chunking documents
|
|
# ---------------------------------------------------------------------------
|
|
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Tools for the agent
|
|
# ---------------------------------------------------------------------------
|
|
@tool
|
|
async def search_knowledge_base(query: str, max_results: int = 5) -> str:
|
|
"""Perform a semantic search in the knowledge base."""
|
|
docs = vector_store.similarity_search(query, k=max_results)
|
|
if not docs:
|
|
return "No relevant documents found."
|
|
return "\n\n".join([f"Title: {doc.metadata.get('title', 'N/A')}\n{doc.page_content}" for doc in docs])
|
|
|
|
@tool
|
|
async def add_to_knowledge_base(content: str, title: str = "Unnamed Document") -> str:
|
|
"""Add a document to the knowledge base after chunking."""
|
|
chunks = text_splitter.split_text(content)
|
|
docs = [Document(page_content=chunk, metadata={"title": title}) for chunk in chunks]
|
|
vector_store.add_documents(docs)
|
|
return f"Document '{title}' added with {len(chunks)} chunks."
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Backend configuration for DeepAgents
|
|
# ---------------------------------------------------------------------------
|
|
backend = CompositeBackend([
|
|
LocalShellBackend(workspace_dir="./workspace"),
|
|
FilesystemBackend(),
|
|
])
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# DeepAgent definition
|
|
# ---------------------------------------------------------------------------
|
|
agent = create_deep_agent(
|
|
model=llm,
|
|
tools=[search_knowledge_base, add_to_knowledge_base],
|
|
backend=backend,
|
|
system_prompt="You are a helpful assistant with access to a knowledge base. Use the provided tools to search and add documents. Respond concisely.",
|
|
)
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Helper function to load all .txt files from a directory into the vector store
|
|
# ---------------------------------------------------------------------------
|
|
async def load_documents_from_directory(directory: str):
|
|
dir_path = Path(directory)
|
|
for txt_file in dir_path.rglob("*.txt"):
|
|
content = txt_file.read_text(encoding="utf-8")
|
|
title = txt_file.stem
|
|
await add_to_knowledge_base(content, title)
|
|
print(f"Loaded {len(list(dir_path.rglob('*.txt')))} documents from {directory}.")
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Interactive CLI client
|
|
# ---------------------------------------------------------------------------
|
|
async def interactive_loop():
|
|
print("Welcome to the RAG Agent CLI. Commands: /add title content | /search query | /quit")
|
|
while True:
|
|
user_input = input("> ")
|
|
if user_input.lower() == "/quit":
|
|
print("Goodbye!")
|
|
break
|
|
if user_input.startswith("/add "):
|
|
try:
|
|
_, rest = user_input.split("/add ", 1)
|
|
title, content = rest.split(" ", 1)
|
|
except ValueError:
|
|
print("Usage: /add title content")
|
|
continue
|
|
human_msg = f"Please add a document titled '{title}' with content: {content}"
|
|
elif user_input.startswith("/search "):
|
|
query = user_input[len("/search "):]
|
|
human_msg = f"Please search for: {query}"
|
|
else:
|
|
print("Unknown command. Use /add, /search, or /quit.")
|
|
continue
|
|
|
|
result = await agent.ainvoke(
|
|
{"messages": [HumanMessage(content=human_msg)]},
|
|
{"configurable": {"thread_id": "session-1"}},
|
|
)
|
|
print(result["messages"][-1].content)
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Main entry point
|
|
# ---------------------------------------------------------------------------
|
|
async def main():
|
|
# Optionally load documents from a directory on startup
|
|
# await load_documents_from_directory("./data")
|
|
await interactive_loop()
|
|
|
|
if __name__ == "__main__":
|
|
asyncio.run(main())
|