​
Setup

Follow the instructions in the MongoDB Setup Guide to get connection string

Install MongoDB packages

pip install "pymongo[srv]"

​
Example

agent_with_knowledge.py
from agno.agent import Agent
from agno.knowledge.pdf_url import PDFUrlKnowledgeBase
from agno.vectordb.mongodb import MongoDb

# MongoDB Atlas connection string
"""
Example connection strings:
"mongodb+srv://<username>:<password>@cluster0.mongodb.net/?retryWrites=true&w=majority"
"mongodb://localhost/?directConnection=true"
"""
mdb_connection_string = ""

knowledge_base = PDFUrlKnowledgeBase(
    urls=["https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf"],
    vector_db=MongoDb(
        collection_name="recipes",
        db_url=mdb_connection_string,
        wait_until_index_ready=60,
        wait_after_insert=300
    ),
)  # adjust wait_after_insert and wait_until_index_ready to your needs

# knowledge_base.load(recreate=True)  # Comment out after first run

agent = Agent(knowledge=knowledge_base, show_tool_calls=True)
agent.print_response("How to make Thai curry?", markdown=True)

Async Support ⚡

MongoDB now supports asynchronous operations for improved performance in production environments.

async_mongodb.py
import asyncio

from agno.agent import Agent
from agno.knowledge.pdf_url import PDFUrlKnowledgeBase
from agno.vectordb.mongodb import MongoDb

# MongoDB Atlas connection string
"""
Example connection strings:
"mongodb+srv://<username>:<password>@cluster0.mongodb.net/?retryWrites=true&w=majority"
"mongodb://localhost:27017/agno?authSource=admin"
"""
mdb_connection_string = "mongodb+srv://<username>:<password>@cluster0.mongodb.net/?retryWrites=true&w=majority"

knowledge_base = PDFUrlKnowledgeBase(
    urls=["https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf"],
    vector_db=MongoDb(
        collection_name="recipes",
        db_url=mdb_connection_string,
    ),
)

# Create and use the agent
agent = Agent(knowledge=knowledge_base, show_tool_calls=True)

if __name__ == "__main__":
    # Comment out after the first run
    asyncio.run(knowledge_base.aload(recreate=False))

    asyncio.run(agent.aprint_response("How to make Thai curry?", markdown=True))

Use aload() and aprint_response() methods with asyncio.run() for non-blocking operations in high-throughput applications.

​
MongoDB Params

ParameterTypeDescriptionDefault
collection_namestrName of the MongoDB collectionRequired
db_urlOptional[str]MongoDB connection string"mongodb://localhost:27017/"
databasestrDatabase name"agno"
embedderOptional[Embedder]Embedder instance for generating embeddingsOpenAIEmbedder()
distance_metricstrDistance metric for similarityDistance.cosine
overwriteboolOverwrite existing collection and index if TrueFalse
wait_until_index_readyOptional[float]Time in seconds to wait until the index is readyNone
wait_after_insertOptional[float]Time in seconds to wait after inserting documentsNone
max_pool_sizeintMaximum number of connections in the connection pool100
retry_writesboolWhether to retry write operationsTrue
clientOptional[MongoClient]An existing MongoClient instanceNone

​
Developer Resources