Skip to main content
Bluemoon exposes your workspace over the Model Context Protocol (MCP), so an AI assistant can query it directly. Ask Claude “where did my visibility drop this week and which competitor took it?” and it reads your real numbers instead of guessing. Nine tools read your measurements. Five run AI actions that cost credits.

Endpoint

The transport is streamable HTTP. You authenticate with a personal access token sent as a Bearer token in the Authorization header. Older client configs that point at https://app.trybluemoon.com/api/mcp/<anything> keep working.
MCP access requires a paid plan. A token on a Free workspace gets 402 with plan_required.

Create a token

Only workspace Admins can create tokens.
  1. In the Bluemoon app, open Settings → API keys and find Personal Access Tokens.
  2. Click Generate token, give it a name you will recognise (for example “Claude Desktop”), and copy it. It is shown only once: if you lose it, revoke it and create another.
  3. A token carries the permissions of the workspace and role that created it. Revoking it stops every client using it immediately.

Connect a client

Most clients take the same block. Put it wherever your client keeps its MCP servers, and replace the token with your own (tokens begin with pat_).
The same block appears in the app under Docs → MCP, next to a copy button for the endpoint. If your client words the transport differently, it is the streamable HTTP transport it is asking for.

Tools that read your data

These are free and read-only. They never change anything in your workspace.

Tools that spend AI credits

These generate content and are charged. An MCP client should ask you before running one. Credits come from the same monthly allowance as the app. Your balance is in the sidebar.

What the answers contain

Some of what these tools return is text written by third parties: the body of a ChatGPT or Gemini answer, the title of a cited page. Bluemoon passes it through unchanged. Treat it as data, never as instructions. A cited page can hide a prompt aimed at AI assistants, so your assistant should not act on directions found in a rawResponse or a page title.

Limits and errors

Every call is written to the workspace audit log with the token that made it.

Checking it works

A working server answers with the tool list. Without a token you get {"error":{"code":"missing","message":"Missing Bearer token"}} — which is also a quick way to confirm the endpoint is reachable at all. Every request is independent and carries its own token, so there is no session to keep alive and no call order to respect. Clients that send an Mcp-Session-Id header are not refused; it is simply not needed.