gsc
MCP server for Google Search Console - query analytics, inspect URLs, manage sitemaps. See Ekamoira for hosted version.
claude mcp add --transport stdio soumyadeep-ux-gsc-mcp-server node /absolute/path/to/gsc-mcp-server/dist/index.js \ --env GSC_LOG_LEVEL="" \ --env GSC_TOKEN_PATH="" \ --env GOOGLE_CLIENT_ID="your-client-id.apps.googleusercontent.com" \ --env GSC_DEFAULT_SITE="" \ --env GOOGLE_CLIENT_SECRET="your-client-secret" \ --env GSC_SERVICE_ACCOUNT_PATH=""
How to use
This MCP server provides a Google Search Console (GSC) integration that exposes the Silence Model Context Protocol (MCP) tools to Claude Desktop, Claude, Cursor, or any MCP client. It lets you query search analytics (clicks, impressions, CTR, and position) across properties, inspect URLs for indexing and crawl data, and manage sitemaps. The server exposes tools such as gsc.list_sites, gsc.search_analytics, gsc.inspect_url, gsc.list_sitemaps, gsc.submit_sitemap, and gsc.delete_sitemap, enabling conversational prompts or scripted automation to interact with GSC data without leaving your MCP client. To connect, configure the MCP client to point at the server's node entrypoint (dist/index.js) and supply Google OAuth credentials or a service account as configured via environment variables. Once authenticated, you can issue prompts like “List all my Search Console sites” or “Submit https://example.com/sitemap.xml to Search Console” and receive structured responses from the MCP server.
How to install
Prerequisites:
- Node.js 20+
- npm (comes with Node.js) or a compatible package manager
- A Google Cloud project with Search Console API enabled
- OAuth 2.0 credentials (Desktop app type) or a service account JSON file
- Install and clone the repository
git clone https://github.com/soumyadeep-ux/gsc-mcp-server.git
cd gsc-mcp-server
npm install
- Create and configure environment
cp .env.example .env
Edit .env to include your Google credentials or service account settings, for example:
GOOGLE_CLIENT_ID=your-client-id.apps.googleusercontent.com
GOOGLE_CLIENT_SECRET=your-client-secret
GSC_SERVICE_ACCOUNT_PATH=/path/to/service-account.json
GSC_TOKEN_PATH=./token.json
GSC_DEFAULT_SITE=https://example.com
GSC_LOG_LEVEL=info
- First-time authentication (OAuth)
npm run auth
This will open a browser to complete Google OAuth consent and save the token locally.
- Run development/build or production workflows
npm run dev # Watch mode with hot reload
npm run build # Compile TypeScript
- Connect via MCP client (example: Claude or Cursor)
- For Claude/Claude Desktop, configure your MCP client to point to the server entry: command: node args: ["/absolute/path/to/gsc-mcp-server/dist/index.js"] env: with your Google credentials as above
- For Cursor, add the server using the MCP integration command, e.g.: npx cursor-mcp add gsc-mcp-server node /absolute/path/to/gsc-mcp-server/dist/index.js
- Optional: run tests or lint
npm run lint
npm run typecheck
npm test
Additional notes
Tips and common issues:
- Use absolute paths in MCP client configurations to avoid file-not-found errors.
- If you encounter token expiration, delete token.json and re-run npm run auth.
- Ensure Google APIs are enabled for the project and that the OAuth client scopes include webmasters or relevant GSC scopes.
- If you switch to a service account, set GSC_SERVICE_ACCOUNT_PATH to the JSON key and ensure the account has access to the GSC properties.
- For Claude Desktop integration, ensure the JSON config in Claude references the exact absolute path to the built server (dist/index.js) and that your environment variables are correctly propagated.
- Logs and debugging: check Claude logs at ~/Library/Logs/Claude/ on macOS and verify your config JSON for valid syntax.
- Environment variables are optional beyond essential credentials; you can omit optional ones or set them to empty strings if not used.
Related MCP Servers
MCPControl
MCP server for Windows OS automation
aws
Talk with your AWS using Claude. Model Context Protocol (MCP) server for AWS. Better Amazon Q alternative.
mcp -typescript
DataForSEO API modelcontextprotocol server
vrchat
This project is a Model Context Protocol (MCP) server for interacting with the VRChat API.
docmole
Dig through any documentation with AI - MCP server for Claude, Cursor, and other AI assistants
google-search-console
It connects directly to your Google Search Console account via the official API, letting you access key data right from AI tools like Claude Desktop or OpenAI Agents SDK and others .