# CrawlCoach developer docs

> Connect an AI agent to CrawlCoach with MCP. Your agent reads your fix list and the pages Google left out, fixes them in your project, and asks CrawlCoach to recheck them with Google.

Each link goes to the page's Markdown copy. For the HTML page, drop `.md` from the URL.

## Docs

- [CrawlCoach developer docs](https://crawlcoach.com/docs.md): Connect an AI agent to CrawlCoach with MCP. Your agent reads your fix list and the pages Google left out, fixes them in your project, and asks CrawlCoach to recheck them with Google.
- [Get started](https://crawlcoach.com/docs/get-started.md): Create a CrawlCoach account, connect Google Search Console, and check your first site.
- [Connect an AI agent](https://crawlcoach.com/docs/connect-an-ai-agent.md): Add the CrawlCoach MCP server to Claude Code, Codex, Claude, VS Code, or Zed. Your agent signs in with your CrawlCoach account and sees only your own sites.
- [MCP tools reference](https://crawlcoach.com/docs/mcp-tools.md): The four CrawlCoach MCP tools and the fix_my_site prompt: inputs, outputs, scopes, and limits.
- [HTTP API reference](https://crawlcoach.com/docs/api.md): The CrawlCoach MCP endpoint over HTTP: how to authenticate, send a JSON-RPC request, and handle errors and rate limits.
- [Plans and limits](https://crawlcoach.com/docs/plans-and-limits.md): What each CrawlCoach plan includes, and the limits on sites, pages, and MCP tools.

## Machine-readable files

- [OpenAPI description](https://crawlcoach.com/openapi.json): the HTTP endpoints, in OpenAPI 3.1.
- [Site summary](https://crawlcoach.com/llms.txt): CrawlCoach's plans and its indexing guides.