Comparison

QuerySail vs mcp-gsc: hosted or self-hosted Search Console MCP

mcp-gsc is the most-starred Search Console MCP server on GitHub. QuerySail is a hosted one. Both let an AI assistant read your Search Console data. They differ in where they run, how you set them up and which AI apps can use them.

Last updated

We make QuerySail. We checked mcp-gsc against its README on October 28, 2026. If something here is wrong or out of date, email support@querysail.com and we will fix it.

Side by side

Featuremcp-gscQuerySail
Where it runsOn your computer, or on a server you run with DockerHosted by QuerySail at one URL
Setupuv, a Google Cloud project with the Search Console API, and an OAuth desktop client or a service accountSign in, approve read-only Search Console access on Google's consent screen and add the URL to your AI app
AI appsApps that start local servers. The README documents Claude Desktop, Cursor and Codex CLIApps that connect to remote MCP servers, including ChatGPT, claude.ai, Claude Code, Codex, Cursor and VS Code
ToolsAbout 15, including search analytics, period comparisons, URL inspection in batches, indexing checks and sitemaps5: performance reports, period comparisons with top gainers and losers, URL inspection for up to 20 URLs, sitemap status and a property list
Changes to Search ConsoleCan submit and delete sitemaps. Tools that add or delete sites are off by defaultNone
Where your Google token is keptOn your computerEncrypted on QuerySail's servers. Your AI app never sees it
Google AnalyticsNot includedIncluded on the same plan
UpdatesYou update it yourselfQuerySail updates it
CostFree (MIT license)$8 per month after a 7-day free trial

Setup, step by step

mcp-gsc, with OAuth

  1. Install uv.
  2. Create a Google Cloud project and turn on the Search Console API.
  3. Create an OAuth client of the type "Desktop app" and download its JSON file.
  4. Add mcp-gsc to your AI app's config file, with the path to the JSON file.
  5. Start your AI app. A browser window opens, and you sign in with Google. mcp-gsc keeps the token on your computer.

QuerySail

  1. Create a QuerySail account. The trial needs no card.
  2. Connect Google under Account → Connections and approve read-only Search Console access.
  3. Add the QuerySail endpoint to your AI app and sign in with your QuerySail account.

Which one to choose

Choose mcp-gsc if

Choose QuerySail if

See every option in the best Google Search Console MCP servers roundup, or read about the QuerySail Google Search Console MCP server.

Questions, answered

Something else? See support.

Is mcp-gsc free?

Yes. mcp-gsc is open source under the MIT license, with about 2,000 stars on GitHub. You pay nothing for the software. You do the setup, and you run the server yourself.

Can I use mcp-gsc with ChatGPT or claude.ai?

Only if you put it on the internet. ChatGPT and claude.ai connect to remote servers only. mcp-gsc can run as a network server with Docker, but then you host it, give it a public HTTPS address and secure it yourself. QuerySail is already on the internet, so ChatGPT and claude.ai can connect to it directly.

Which one has more tools?

mcp-gsc has more tools, and it can submit and delete sitemaps. QuerySail has five tools that cover performance reports, comparisons, URL inspection and sitemap status. It cannot change anything in Search Console.

Can I use both?

Yes. They do not conflict. For example, you can run mcp-gsc in Claude Desktop or Cursor and use QuerySail in ChatGPT or on claude.ai, where a local server cannot connect.