MCP
Rightbrain speaks the Model Context Protocol (MCP), the open standard for connecting AI applications to external tools. This page covers the outbound direction: turning your Rightbrain tasks and agents into MCP tools that external clients discover and run.
Two directions, one protocol. This page is about exposing Rightbrain as an MCP server so tools like Claude and Cursor can call your work. For the other direction, where a Rightbrain agent connects out to an external MCP server as a client, see Connections.
What you get
When you expose a task over MCP, it becomes discoverable and runnable by any MCP-compatible client, including Claude.ai, Cursor, and Windsurf. The client sees your task as a native tool. It calls the tool, Rightbrain runs the task on your configured model with your prompt and schema, and returns the structured result.
Expose a task over MCP
A task is served over MCP once it’s flagged for agent exposure and you have an API key for clients to authenticate with. Both are set through the API (or the dashboard).
Flag the task for exposure
Set the task’s exposed_to_agents flag. The MCP gateway advertises every exposed task in the project as a callable tool.
The MCP gateway
Under the hood, exposed tasks and agents are served through Rightbrain’s MCP gateway over Streamable HTTP. Clients that let you add a server by URL connect to the project-scoped gateway endpoint:
Authenticate with the API key you selected when enabling MCP, sent as a bearer token:
The gateway advertises each enabled task as a tool. Tool calls run the task and stream results back to the client.