1  """LanceDB Vector Store - Agent-First Example"""
 2  from praisonaiagents import Agent
 3  
 4  # Agent-first approach: use knowledge parameter with LanceDB
 5  agent = Agent(
 6      name="Assistant",
 7      instructions="You are a helpful assistant with access to documents.",
 8      knowledge={"sources": ["./docs/guide.pdf"], "vector_store": {"provider": "lancedb", "path": "/tmp/lancedb_test"}}
 9  )
10  
11  # Chat - agent uses knowledge for RAG
12  response = agent.chat("What information do you have?")
13  print(f"Response: {response}")
14  
15  print("PASSED: LanceDB with Agent")
16  
17  # --- Advanced: Direct LanceDB Usage ---
18  # import lancedb
19  # db = lancedb.connect("/tmp/lancedb_test")
20  # data = [{"id": "1", "text": "ML is AI", "vector": [0.1] * 128}]
21  # table = db.create_table("demo", data)