Implementation:Cohere ai Cohere python Connector Model

Knowledge Sources	Cohere Python SDK
Domains	SDK, Connectors, RAG
Last Updated	2026-02-15 14:00 GMT

Overview

Connector is a Pydantic model representing a data source connector entity that integrates external data sources with the Cohere chat endpoint for retrieval-augmented generation.

Description

The Connector model represents a registered data source connector in the Cohere platform. Connectors allow you to integrate external data sources with the /chat endpoint to create grounded generations with citations to the data source. Each connector has:

• A unique id that is automatically created from the connector name upon registration.
• An organization_id automatically set to the organization of the user who created it.
• A human-readable name and optional description.
• A url pointing to the search endpoint that will be queried for documents.
• Timestamp fields created_at and updated_at tracking lifecycle.
• An excludes list of fields to exclude from the prompt while keeping them in the document.
• Authentication configuration via auth_type (oauth or service_auth), oauth (ConnectorOAuth), and auth_status (ConnectorAuthStatus).
• An active flag indicating whether the connector is enabled.
• A continue_on_failure flag controlling whether chat requests proceed if this connector fails.

Usage

Use Connector when working with the Cohere connectors API to create, list, update, or delete data source connectors. This model is returned by connector management endpoints and referenced when configuring connectors for RAG-enabled chat requests.

Code Reference

Source Location

• Repository: Cohere Python SDK
• File: src/cohere/types/connector.py

Signature

class Connector(UncheckedBaseModel):
    id: str = pydantic.Field()
    organization_id: typing.Optional[str] = pydantic.Field(default=None)
    name: str = pydantic.Field()
    description: typing.Optional[str] = pydantic.Field(default=None)
    url: typing.Optional[str] = pydantic.Field(default=None)
    created_at: dt.datetime = pydantic.Field()
    updated_at: dt.datetime = pydantic.Field()
    excludes: typing.Optional[typing.List[str]] = pydantic.Field(default=None)
    auth_type: typing.Optional[str] = pydantic.Field(default=None)
    oauth: typing.Optional[ConnectorOAuth] = pydantic.Field(default=None)
    auth_status: typing.Optional[ConnectorAuthStatus] = pydantic.Field(default=None)
    active: typing.Optional[bool] = pydantic.Field(default=None)
    continue_on_failure: typing.Optional[bool] = pydantic.Field(default=None)

Import

from cohere.types import Connector

I/O Contract

Fields

Field	Type	Required	Description
id	str	Yes	The unique identifier of the connector, auto-created from its name.
organization_id	Optional[str]	No	The organization to which this connector belongs.
name	str	Yes	A human-readable name for the connector.
description	Optional[str]	No	A description of the connector.
url	Optional[str]	No	The URL of the connector used to search for documents.
created_at	datetime	Yes	The UTC time at which the connector was created.
updated_at	datetime	Yes	The UTC time at which the connector was last updated.
excludes	Optional[List[str]]	No	A list of fields to exclude from the prompt (fields remain in the document).
auth_type	Optional[str]	No	The authentication type used by the connector. Possible values: "oauth", "service_auth".
oauth	Optional[ConnectorOAuth]	No	The OAuth 2.0 configuration for the connector.
auth_status	Optional[ConnectorAuthStatus]	No	The OAuth status for the requesting user. One of "valid", "expired", or empty.
active	Optional[bool]	No	Whether the connector is active or not.
continue_on_failure	Optional[bool]	No	Whether a chat request should continue if the request to this connector fails.

Usage Examples

from cohere.types import Connector

# List all connectors
connectors = client.connectors.list()
for connector in connectors.connectors:
    print(f"ID: {connector.id}")
    print(f"Name: {connector.name}")
    print(f"URL: {connector.url}")
    print(f"Active: {connector.active}")
    print(f"Auth type: {connector.auth_type}")
    print(f"Created: {connector.created_at}")

# Get a specific connector by ID
connector = client.connectors.get(id="my-connector-id")
print(f"Connector: {connector.name}")
print(f"Description: {connector.description}")
print(f"Continue on failure: {connector.continue_on_failure}")

# Use a connector in a chat request for RAG
response = client.chat(
    message="What are the latest updates?",
    connectors=[{"id": connector.id}],
)

Related Pages

• Environment:Cohere_ai_Cohere_python_Python_SDK_Runtime