Teams with Knowledge

Attach a knowledge base the team coordinator can search.

Pass knowledge to a team to register its search_knowledge_base tool.

from pathlib import Path

from agno.agent import Agent
from agno.knowledge.embedder.openai import OpenAIEmbedder
from agno.knowledge.knowledge import Knowledge
from agno.models.openai import OpenAIResponses
from agno.team import Team
from agno.tools.hackernews import HackerNewsTools
from agno.vectordb.lancedb import LanceDb

cwd = Path(__file__).parent
tmp_dir = cwd.joinpath("tmp")
tmp_dir.mkdir(parents=True, exist_ok=True)

agno_docs_knowledge = Knowledge(
    vector_db=LanceDb(
        uri=str(tmp_dir.joinpath("lancedb")),
        table_name="agno_docs",
        embedder=OpenAIEmbedder(id="text-embedding-3-small"),
    ),
)

hackernews_agent = Agent(
    name="HackerNews Agent",
    role="Search HackerNews for tech news",
    model=OpenAIResponses(id="gpt-5.2"),
    tools=[HackerNewsTools()],
    instructions=["Always include sources"],
)

team_with_knowledge = Team(
    name="Team with Knowledge",
    members=[hackernews_agent],
    model=OpenAIResponses(id="gpt-5.2"),
    knowledge=agno_docs_knowledge,
    show_members_responses=True,
    markdown=True,
)

if __name__ == "__main__":
    agno_docs_knowledge.insert(url="https://docs.agno.com/llms-full.txt")
    team_with_knowledge.print_response("Tell me about the Agno framework", stream=True)

uv pip install -U agno lancedb openai

The team coordinator can search the attached knowledge base and delegate other work to members. Team.search_knowledge defaults to True.

When multiple Knowledge instances share one vector database, set isolate_vector_search=True (and give each instance a name) to keep retrieval scoped to each instance’s own content.

Next Steps

Task	Guide
Add web search to a team	Team with Knowledge Base
Give members separate knowledge bases	Distributed RAG with LanceDB
Configure ingestion and retrieval	Knowledge Overview