Skip to main content
Build agents that search and retrieve information from your documents, databases, and APIs. 
from agno.agent import Agent
from agno.knowledge.knowledge import Knowledge
from agno.vectordb.pgvector import PgVector

knowledge = Knowledge(
    vector_db=PgVector(table_name="documents", db_url="postgresql://..."),
)

# Add documents from any source
knowledge.add_content(url="https://example.com/docs.pdf")
knowledge.add_content(path="./local_docs/")

# Create an agent with knowledge
agent = Agent(knowledge=knowledge, search_knowledge=True)
agent.print_response("What does the documentation say about authentication?")

​
Knowledge Categories

Vector Databases

25+ vector databases including PgVector, Pinecone, Qdrant, Weaviate.

Embedders

29 embedder options from OpenAI, Cohere, HuggingFace, and more.

Document Readers

PDF, CSV, JSON, Markdown, PowerPoint, and web content readers.

Chunking

Semantic, fixed-size, and custom chunking strategies.

​
Quick Examples

​
Basic Knowledge Base

cookbook/07_knowledge/vector_db/pgvector/pgvector_db.py
from agno.agent import Agent
from agno.knowledge.knowledge import Knowledge
from agno.vectordb.pgvector import PgVector

vector_db = PgVector(
    table_name="recipes",
    db_url="postgresql+psycopg://ai:ai@localhost:5532/ai",
)

knowledge = Knowledge(
    name="Recipe Knowledge",
    vector_db=vector_db,
)

# Add content from URL
knowledge.add_content(
    name="Thai Recipes",
    url="https://example.com/recipes.pdf",
)

agent = Agent(
    knowledge=knowledge,
    search_knowledge=True,
)

agent.print_response("How do I make pad thai?", markdown=True)

​
Multiple Sources

cookbook/07_knowledge/basic_operations/sync/04_from_multiple.py
from agno.knowledge.knowledge import Knowledge
from agno.vectordb.pgvector import PgVector

knowledge = Knowledge(
    vector_db=PgVector(table_name="docs", db_url="postgresql://..."),
)

# Add from multiple sources
knowledge.add_content(path="./documentation/")
knowledge.add_content(url="https://docs.example.com/api.pdf")
knowledge.add_content(topic="Python best practices")

​
Async Operations

cookbook/07_knowledge/basic_operations/async/01_from_path.py
import asyncio
from agno.knowledge.knowledge import Knowledge
from agno.vectordb.pgvector import PgVector

async def build_knowledge():
    knowledge = Knowledge(
        vector_db=PgVector(table_name="docs", db_url="postgresql://..."),
    )

    # Async content loading
    await knowledge.add_content_async(path="./large_docs/")

    return knowledge

knowledge = asyncio.run(build_knowledge())

​
Run Examples

git clone https://github.com/agno-agi/agno.git
cd agno/cookbook/07_knowledge

# PgVector example
python vector_db/pgvector/pgvector_db.py

# Multiple sources
python basic_operations/sync/04_from_multiple.py

# Embedders
python embedders/openai_embedder.py