Mongo Agent Storage

Phidata supports using Mongo as a storage backend for Agents using the MongoAgentStorage class.

Usage

You need to provide either db_url or client. The following example uses db_url.

storage.py

from phi.storage.agent.mongodb import MongoAgentStorage

db_url = "mongodb://ai:ai@localhost:27017/phi"

# Create a storage backend using the Mongo database
storage = MongoAgentStorage(
    # store sessions in the agent_sessions collection
    collection_name="agent_sessions",
    db_url=db_url,
)

# Add storage to the Agent
agent = Agent(storage=storage)

Params

Parameter	Type	Default	Description
`collection_name`	`str`	-	Name of the collection to be used.
`db_url`	`Optional[str]`	`None`	Database URL, if provided.
`db_name`	`str`	`"phi"`	Database Name.
`client`	`Optional[MongoClient]`	`None`	MongoDB client, if provided.

​Usage

​Params

Usage

Params