Multi-Source RAG: Combining Different Content Types

In production, agents often need knowledge from multiple sources: PDFs, web pages, text snippets, and databases.

multi_source_rag.py

"""
Multi-Source RAG: Combining Different Content Types
====================================================
In production, agents often need knowledge from multiple sources:
PDFs, web pages, text snippets, and databases.

This example loads content from different source types into the same
knowledge base, demonstrating insert_many with mixed sources.

See also: 02_knowledge_lifecycle.py for managing content over time.
"""

import asyncio

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.vectordb.qdrant import Qdrant
from agno.vectordb.search import SearchType

# ---------------------------------------------------------------------------
# Setup
# ---------------------------------------------------------------------------

qdrant_url = "http://localhost:6333"

knowledge = Knowledge(
    vector_db=Qdrant(
        collection="multi_source_rag",
        url=qdrant_url,
        search_type=SearchType.hybrid,
        embedder=OpenAIEmbedder(id="text-embedding-3-small"),
    ),
)

agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    knowledge=knowledge,
    search_knowledge=True,
    markdown=True,
)

# ---------------------------------------------------------------------------
# Run Demo
# ---------------------------------------------------------------------------

if __name__ == "__main__":

    async def main():
        # Load multiple sources in a single batch call
        await knowledge.ainsert_many(
            [
                {
                    "name": "Candidate Resume",
                    "path": "cookbook/07_knowledge/testing_resources/cv_1.pdf",
                    "metadata": {"source": "resume", "department": "engineering"},
                },
                {
                    "name": "Thai Recipes",
                    "url": "https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf",
                    "metadata": {"source": "web", "topic": "cooking"},
                },
                {
                    "name": "Company Policy",
                    "text_content": "All employees must complete security training annually. "
                    "Remote work requires VPN access. Expenses over $500 need manager approval.",
                    "metadata": {"source": "internal", "topic": "policy"},
                },
            ]
        )

        print("\n" + "=" * 60)
        print("Query across multiple sources")
        print("=" * 60 + "\n")

        agent.print_response("What skills does Jordan Mitchell have?", stream=True)

        print("\n" + "=" * 60)
        print("Agent searches the same knowledge base for different topics")
        print("=" * 60 + "\n")

        agent.print_response("What is the expense approval policy?", stream=True)

    asyncio.run(main())

Run the Example

1. Set up your virtual environment

   Mac

   uv venv --python 3.12
   source .venv/bin/activate
   

   uv venv --python 3.12
   .venv\Scripts\activate
   
2. Install dependencies

   uv pip install -U agno fastembed openai pypdf qdrant-client
   
3. Export your OpenAI API key

   Mac/Linux

   export OPENAI_API_KEY="your_openai_api_key_here"
   

   $Env:OPENAI_API_KEY="your_openai_api_key_here"
   
4. Run Qdrant

   docker run -d --name qdrant -p 6333:6333 qdrant/qdrant:latest
   
5. Clone Agno

   Clone the repository and run the remaining commands from its root:

   git clone --branch v2.7.4 --depth 1 https://github.com/agno-agi/agno.git
   cd agno
   
6. Run the example

   Run the example from the repository root:

   python cookbook/07_knowledge/03_production/01_multi_source_rag.py
   

Full source: cookbook/07_knowledge/03_production/01_multi_source_rag.py