Google Search Console MCP Server (read-only by default)

An MCP server for Google Search Console, using the Search Console API (Search Analytics, Sites and Sitemaps) and the URL Inspection API. Use it from Claude, ChatGPT, Cursor, Codex, Gemini or any MCP client to report on organic search performance, compare periods, check sitemaps and inspect how Google indexes your pages.

• search analytics: clicks, impressions, CTR and position by query, page, country, device, date or search appearance, for web, image, video, news, Discover and Google News

• period comparisons: this period against the last one or last year, with winners and losers, and brand vs non-brand totals

• page and query reports: the queries a page ranks for, and the pages that compete for a query

• URL Inspection: index status, coverage reason, Google's canonical against yours, last crawl, rich results; one URL or a paced batch of up to 200

• sitemaps: status, errors and warnings; submit and delete with --allow-write

• built-in Google sign-in: no gcloud needed

Part of the AI Shopping Feeds MCP family: Google Ads · Google Merchant Centre · Google Analytics · YouTube Studio · Google Search Console.

Let your agent set it up

Paste this into Claude Code, Codex, Cursor or any agent that can read web pages and edit files:

Set up the Google Search Console MCP server for me by following https://raw.githubusercontent.com/AI-Shopping-Feeds/search-console-mcp/main/skills/search-console-mcp/SKILL.md

The agent checks whether it's already connected, installs the runner if you agree, asks for your OAuth client file and has you sign in to Google in your browser, adds the server to your client's config without touching your other servers, and checks it works by calling list_sites.

The same file is an Agent Skill, so your agent can keep it as a guide to using the tools well. For Claude Code, copy the folder into your skills:

git clone --depth 1 https://github.com/AI-Shopping-Feeds/search-console-mcp.git /tmp/search-console-mcp
cp -r /tmp/search-console-mcp/skills/search-console-mcp ~/.claude/skills/

Other agents that support skills load the same skills/search-console-mcp folder from their own skills directory.

Related MCP server: Google Search Console MCP Server

What it can do

12 tools: 10 read tools, plus 2 write tools when the server runs with --allow-write.

Read

Write

Only listed when the server starts with --allow-write.

Nothing else writes. Search Console's API can't change settings, remove URLs or request indexing.

Safety

• The server is read-only unless it starts with --allow-write. Without it, it asks Google only for the webmasters.readonly scope.

• Tools carry MCP titles and annotations; delete_sitemap is marked destructive so clients can ask first.

• batch_inspect_urls keeps to Google's URL Inspection quota (600 a minute and 2,000 a day per property) and stops cleanly if Google says the quota is used up.

• The sign-in token is saved with 0600 permissions in your user config directory, never in the project.

Supported clients

Setup

1. Install uv

The server runs with uvx from uv (brew install uv on macOS) and needs Python 3.10 or later. It installs straight from this repository:

uvx --from git+https://github.com/AI-Shopping-Feeds/search-console-mcp.git search-console-mcp

Add --allow-write after search-console-mcp to enable the sitemap write tools (here and in the client configs below). Pin a release by adding @vX.Y.Z to the URL.

2. Get credentials

1. Google Cloud project: enable the Google Search Console API and create an OAuth client of type Desktop app. Download its JSON.

2. Sign in as a Google user with access to your properties. Your browser opens Google's sign-in page; the token is saved to ~/.config/search-console-mcp/token.json (Windows: %APPDATA%\search-console-mcp\token.json). Add --allow-write for the write scope.

   uvx --from git+https://github.com/AI-Shopping-Feeds/search-console-mcp.git \
     search-console-mcp auth --client-secrets /abs/path/client_secret.json

Service accounts and Application Default Credentials also work: docs/credentials.md.

3. Add it to your client

Keep your other servers in each file. With the built-in sign-in, no environment variables are needed.

Claude Code

claude mcp add --scope user search-console \
  -- uvx --from git+https://github.com/AI-Shopping-Feeds/search-console-mcp.git search-console-mcp

Claude Desktop, Cursor, Gemini CLI, Windsurf (and VS Code under servers)

{
  "mcpServers": {
    "search-console": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/AI-Shopping-Feeds/search-console-mcp.git", "search-console-mcp"]
    }
  }
}

Desktop apps often can't find uvx on the PATH; if the server doesn't start, use the full path from which uvx as the command.

Codex (~/.codex/config.toml)

[mcp_servers.search-console]
command = "uvx"
args = ["--from", "git+https://github.com/AI-Shopping-Feeds/search-console-mcp.git", "search-console-mcp"]

4. Check it works

Restart or reload the client, then ask it to "list my Search Console properties". It should call list_sites.

Configuration

Tool groups keep the tool list short when you only need part of it. list_sites and get_site are always registered, and write tools still need --allow-write. An unknown group stops the server at startup.

Example prompts

Compare the last 3 months with the 3 months before: clicks, impressions, CTR and position, split into brand and non-brand for "example".

Which queries lost the most clicks this month compared with last month?

Which pages compete for "running shoes", and which queries does https://www.example.com/collections/shoes rank for?

Inspect the home page and these product URLs. Are any not indexed, or is Google choosing a different canonical?

More docs

• Credentials: the built-in sign-in, service accounts and Application Default Credentials

• Agent skill: setup and usage guide for agents

Contributing and licence

Issues and pull requests are welcome on this repository. See CONTRIBUTING.md.

Licensed under the Apache License 2.0.

• The server sends your Search Console data to the agent or LLM you connect it to.

Try out AI Shopping Feeds, our shopping feed management tool (including Google Shopping feeds).