connector
A connector links an AI tool to the data or functions of an external app or service.
A connector is a software component that lets an AI tool access data or invoke functions in an external app or service. It may use the service’s API or a standard such as the Model Context Protocol (MCP). MCP specifies a way to connect tools and services; a connector implements a connection to a particular service. A connector built on MCP is also called an MCP connector; Anthropic uses the name MCP connector for the Claude API feature that connects directly to remote MCP servers.
An AI agent can use a connector to retrieve information or perform actions in another service. What it can do depends on the connected service and the access permissions granted. Some AI chat services, including Anthropic’s Claude, use the name Connectors for features that link them to external apps, data and tools.
This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.
Articles covering this entry
Why Claude Code needs a video connector
Claude Code walks the user through most installations and handles many of them itself, but third-party connectors have to be linked by hand.
Today a user creates an agent through a chat-style interface and picks the work apps it should connect to from a library of prebuilt connectors.
Is there a plugin or connector?
A connector might let it read reference files in Google Drive or write an output there; a plugin may bundle tools, prompts, and authentication for a service.
Meta has opened Muse, its personal AI assistant, to third-party services through Muse Connectors.
In vidIQ account settings, open the MCP menu under API Keys and copy the connector URL.