wtf is search-console-mcp?

search-console-mcp lets you talk to your Google Search Console data through an AI assistant like Claude or Cursor, instead of digging through dashboards yourself. Google Search Console is Google's free tool for seeing how your website performs in search results, showing things like which search queries bring people to your site and whether your pages are properly indexed. This project connects that data to an AI chat assistant using MCP, short for Model Context Protocol, which is a standard way for AI assistants to plug into outside tools and data sources. What sets this project apart is how simple the setup is. Instead of requiring you to create a Google Cloud account, enable APIs, and manage credential files, which is the normal way other similar tools work, this one just needs you to run a single command and sign in with your Google account in your browser. The whole process takes about thirty seconds, and the login tokens are stored only on your own computer, never passing through anyone else's servers. The tool is also read-only by design, meaning it can see your search data but cannot make any changes to your website or account settings. Once connected, you can ask plain questions like which pages lost the most clicks this month, which search queries are close to reaching the first page of results, or whether a specific page is indexed by Google. The project also ships five ready-made analyses you can run, covering an overall health checkup, finding pages that compete with each other for the same search terms, spotting pages just outside the top results, diagnosing sudden traffic drops, and auditing indexing status. Setup requires Node.js, since the tool runs through npx, a command that runs Node packages without a separate install step. The README emphasizes that its privacy claims can be checked directly, since the authentication code is only about two hundred lines and is included in the project's source files.