Files
cucumbers-solutions/solutions/6a02e23da6fe2e4ac16acf65/solution.py
T

92 lines
3.7 KiB
Python

from langchain_ollama import ChatOllama, OllamaEmbeddings
from langchain_qdrant import QdrantVectorStore
from qdrant_client import QdrantClient
from qdrant_client.http.models import Distance, VectorParams
from langchain_core.documents import Document
from langchain.tools import tool
from langchain.agents import create_agent
import os
# ---------- LLM and embeddings ----------
llm = ChatOllama(model="llama3", temperature=0.2)
embeddings = OllamaEmbeddings(model="nomic-embed-text")
# ---------- Qdrant client & collection ----------
client = QdrantClient(":memory:")
collection_name = "knowledge_base"
client.create_collection(
collection_name=collection_name,
vectors_config=VectorParams(size=embeddings.embedding_size, distance=Distance.COSINE),
)
vector_store = QdrantVectorStore(client=client, collection_name=collection_name, embedding=embeddings)
# ---------- Text splitter ----------
from langchain_text_splitters import RecursiveCharacterTextSplitter
splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
# ---------- Tools ----------
@tool
def search_knowledge_base(query: str, max_results: int = 5) -> str:
"""Search the knowledge base for relevant documents."""
results = vector_store.similarity_search_with_score(query, k=max_results)
if not results:
return "No relevant documents found."
out_lines = []
for doc, score in results:
title = doc.metadata.get("title", "Untitled")
content = doc.page_content[:200] + ("..." if len(doc.page_content) > 200 else "")
out_lines.append(f"Title: {title}\nScore: {score:.4f}\nContent: {content}")
return "\n\n".join(out_lines)
@tool
def add_to_knowledge_base(content: str, title: str = "Untitled") -> str:
"""Add a new document to the knowledge base."""
chunks = splitter.split_text(content)
docs = [Document(page_content=chunk, metadata={"title": title}) for chunk in chunks]
vector_store.add_documents(docs)
return f"Added {len(chunks)} chunks under title '{title}'."
# ---------- Agent ----------
system_prompt = """
You are an assistant that can search and add documents to a knowledge base.
Use the tools `search_knowledge_base` and `add_to_knowledge_base` as needed.
Respond with plain text. Do not mention tool usage explicitly unless required by the user.
"""
agent = create_agent(
model=llm,
tools=[search_knowledge_base, add_to_knowledge_base],
system_prompt=system_prompt,
)
# ---------- CLI ----------
def main():
print("RAG Agent CLI. Commands: /add <title> | <content>, /search <query>, /quit")
while True:
user_input = input("\nYou: ").strip()
if not user_input:
continue
if user_input.lower() in ("exit", "quit", "/quit"):
print("Goodbye!")
break
if user_input.startswith("/add "):
try:
_, rest = user_input.split(maxsplit=1)
title, content = rest.split("|", 1)
title = title.strip()
content = content.strip()
result_msg = add_to_knowledge_base(content=content, title=title)
print(f"Bot: {result_msg}")
except ValueError:
print("Bot: Usage /add <title> | <content>")
elif user_input.startswith("/search "):
query = user_input[len("/search "):].strip()
result_msg = search_knowledge_base(query=query, max_results=5)
print(f"Bot:\n{result_msg}")
else:
# Regular chat
response = agent.invoke({"messages": [{"role": "human", "content": user_input}]})
ai_message = response["messages"][-1]
print(f"Bot: {ai_message.content}")
if __name__ == "__main__":
main()