Google Search Console MCP Server is a Model Context Protocol (MCP) server used to connect Google Search Console with Claude Desktop, Cursor, Windsurf, and other MCP-compatible clients. Users can analyze SEO data, inspect URLs, discover keyword opportunities, and track search performance through natural language. It is officially stated that no Google Cloud setup is required; after installation, users authenticate with a Google account and can begin querying Search Console data in about 2 minutes.
Purpose This tool is aimed at SEO and website operations scenarios, with main uses including: querying search analytics data, checking URL indexing status, discovering keyword opportunities, tracking SEO performance, and conducting branded and non-branded traffic analysis.
Main capabilities According to official information, the server provides 13 SEO tools, with capabilities covering: SEO performance monitoring (tracking changes in clicks, impressions, CTR, and average ranking over time); keyword research and opportunity discovery (finding keywords that rank but have high impressions and low CTR); content optimization (identifying pages that need improvement based on search data); technical SEO audits (checking URL indexing status, sitemap health, and mobile usability); comparative analysis (comparing performance across different periods and identifying trends and declines); reporting (exporting data for SEO reports and dashboards); brand monitoring (tracking branded and non-branded search traffic).
How to use
- After installation, run gsc-mcp-auth for one-time authentication: the browser opens the Google login page, log in with a Google account and grant read-only access to Search Console; the token is stored locally in ~/.gsc-mcp/tokens/. It is officially stated that multiple accounts are supported.
- Add the server in the Claude Desktop configuration file (on macOS the path is ~/Library/Application Support/Claude/claude_desktop_config.json), then restart Claude Desktop to use it. It can also be used with other MCP-compatible clients, such as Cursor, Windsurf, VS Code + Cline, etc.; simply point the MCP client to the corresponding entry point (the specific command is not fully provided in the materials).
Example usage Four examples are officially provided:
- "Show the top 10 pages by clicks in the last 30 days": call get_top_pages, passing in the site URL, a date range for the last 30 days, sortBy: "clicks", limit: 10, and return a list of pages ranked by clicks along with clicks, impressions, CTR, and average ranking.
- "Find keywords on example.com with high impressions but low CTR": call find_keyword_opportunities, use a 28-day date range, filter for queries with more than 100 impressions, ranking in the top 20, and CTR below 3%, and return keywords whose clicks can be improved by optimizing titles, meta descriptions, or content.
- "Check https://example.com/blog/my-new-post "Whether it is indexed": call inspect_url, which returns the index status (indexed, crawled but not indexed, not found, etc.), the most recent crawl date, mobile usability status, and detected rich media results, and identifies issues blocking indexing such as robots.txt blocks, noindex tags, or crawl errors.
- "Compare example.com's search performance this month vs. last month": call compare_performance, which returns the queries with the largest increase or decrease in clicks along with the change values for clicks, impressions, CTR, and ranking, highlighting trending keywords and declining pages.
Use cases Suitable for SEO performance monitoring, keyword research and opportunity discovery, content optimization decisions, technical SEO audits, competitor/period comparison analysis, SEO report export, and brand traffic monitoring, etc.
Authentication and data security This package has built-in OAuth credentials, so there is no need to create a Google Cloud project. During authentication, the browser opens the Google sign-in page, and the user logs in and grants read-only access to Search Console data; the token is stored only locally in ~/.gsc-mcp/tokens/. The official privacy statement says: this extension runs locally, OAuth tokens are stored locally, and all Search Console data is queried directly through the Google API without going through any intermediary, and no data is collected, stored, or transmitted to third parties. The OAuth credentials are used only to identify the application; each user authenticates separately and can access only their own Search Console data. If you want to use your own OAuth credentials, there are two optional methods: A. Use environment variables; B. Save the credentials JSON to ~/.gsc-mcp/credentials.json. Steps to create your own credentials: go to the Google Cloud Console, create or select a project, enable the Google Search Console API, go to "Credentials > Create Credentials > OAuth 2.0 Client ID", select "Desktop app" as the application type, and download the credentials JSON file.
Limitations and requirements Usage requires a Google account with access to Google Search Console. The materials do not provide specific fees, pricing, platform limitations, or performance data, so no relevant information is available at this time.
