An MCP server publishes a set of tools, each with a name, description and input schema. An AI client connects to it and calls those tools on the user's behalf.
In VisibilityKit Crawler, the MCP server is built into the app and runs locally on your machine. There is no hosted MCP endpoint.
A built-in server means your AI assistant configures and starts a real crawl, not a saved profile it has to pick from, and the desktop window fills in live as it runs.
VisibilityKit Crawler's MCP server exposes 36 tools, including starting and pausing crawls, listing issues, reading pages and links, comparing crawls, saving snapshots, exports and fix checks. It is included on every plan and works with any LLM or AI client that can connect to and call an MCP server.
Ask your AI assistant to start a full crawl of your site. The Crawler window opens the crawl and fills in live, and the assistant can then list the top issues and re-check a page after you ship a fix.
No. It ships with the app, so there is nothing separate to install or keep running.
Yes.
An open standard that lets AI assistants call external tools and read data through a common interface.
Re-checking a page after a fix ships to confirm, with evidence, that the issue is actually resolved.
Comparing two crawls of the same site to see what changed between them.
VisibilityKit Crawler's workspace for turning audit findings into assigned, tracked work.
A desktop SEO crawler for macOS, Windows and Linux. Find what is broken, assign the work, then verify every fix. Crawl data stays on your machine.
Get Crawler free