The MetaMonster API

Everything you see in the MetaMonster dashboard is available over HTTP: sites, pages, content, analyses, recommendations, drafts, and keyword metrics. That means you can script bulk work, wire MetaMonster into your own tooling, or let an AI agent drive it for you.

This guide is the orientation. The full reference lives at docs.metamonster.ai, with a guide for every concept and documentation for every endpoint.

Just want to use MetaMonster from a chat? You don’t need an API key for that. See Connect MetaMonster to Claude or Connect MetaMonster to ChatGPT.

Create an API key

The REST API, and coding tools that connect to the MCP server, authenticate with an API key. Go to Settings → API Keys and click Create key.

Keys start with mm_ and are shown once, at creation. Store the key somewhere safe and treat it like a password: it can read and modify your organization’s SEO data. If a key leaks, revoke it on the same page and create a new one.

The REST API

The base URL is:

https://app.metamonster.ai/api/v1

Send your key as a bearer token. The quickest way to confirm everything works:

curl https://app.metamonster.ai/api/v1/me \
  -H "Authorization: Bearer mm_YOUR_API_KEY"

A valid key returns its name and scopes. From there you can list sites, pull page reports, stage content edits, trigger analyses, and export keyword data. Start with the getting started guide, then work from the endpoint reference. There’s also an OpenAPI spec if you’d rather generate a client.

Coding tools

Coding tools like Cursor and Windsurf can use MetaMonster through its MCP server, which wraps the API in tools built for agents. Add it to mcp.json with your API key:

{
  "mcpServers": {
    "metamonster": {
      "url": "https://app.metamonster.ai/mcp",
      "headers": { "Authorization": "Bearer mm_YOUR_KEY" }
    }
  }
}

For Claude Code, see Connect MetaMonster to Claude. The MCP server doc lists every tool and the error shapes.

Costs and limits

API and MCP calls draw from the same plan allowances as the dashboard: working on a page uses a page credit and counts toward your active sites, exactly as if you’d done it in the dashboard. Rate limits and request conventions are covered in the conventions doc.