mcp-seo-audit
SEO audit and Google Search Console MCP server with 23 tools. Search analytics, URL inspection, Indexing API, Core Web Vitals (CrUX), striking distance keywords, keyword cannibalization detection, branded query analysis, and automated site audits.
mcp-seo-audit
A Model Context Protocol (MCP) server for SEO auditing, Google Search Console analytics, URL indexing, Core Web Vitals, and site analysis. Works with Claude Code, Claude Desktop, Cursor, and any MCP-compatible client.
Forked from AminForou/mcp-gsc and rewritten with 23 tools, bug fixes, and a full test suite.
What It Does
| Category | Tools | Description |
|---|---|---|
| Property Management | list_properties, add_site, delete_site | List, add, and remove GSC properties |
| Search Analytics | get_search_analytics, get_advanced_search_analytics, get_performance_overview, get_search_by_page_query, compare_search_periods | Query clicks, impressions, CTR, position with filtering, dimensions, and period comparison |
| URL Inspection | inspect_url, batch_inspect_urls | Check indexing status, crawl info, canonical, robots for one or many URLs |
| Indexing API | request_indexing, request_removal, check_indexing_notification, batch_request_indexing | Submit/remove URLs from Google's index via the Indexing API |
| Sitemaps | get_sitemaps, submit_sitemap, delete_sitemap | List, submit, and delete sitemaps |
| Core Web Vitals | get_core_web_vitals | LCP, FID, CLS, INP, TTFB via the Chrome UX Report (CrUX) API |
| SEO Analysis | find_striking_distance_keywords, detect_cannibalization, split_branded_queries | Find keywords at positions 5-20, detect pages competing for the same query, split branded vs non-branded traffic |
| Site Audit | site_audit | All-in-one report: sitemap health, indexing status, canonical mismatches, performance summary |
| Auth | reauthenticate | Switch Google accounts by clearing cached OAuth tokens |
23 tools total.
Setup
1. Google API Credentials
OAuth (recommended)
- Go to Google Cloud Console
- Enable the Search Console API and Web Search Indexing API
- Create an OAuth 2.0 Client ID (Desktop app)
- Download
client_secrets.json
Service Account
- Create a service account in Google Cloud Console
- Download the JSON key file
- Add the service account email to your GSC properties
2. Install
git clone https://github.com/GiorgiKemo/mcp-gsc.git
cd mcp-gsc
python -m venv .venv
# Activate:
# macOS/Linux: source .venv/bin/activate
# Windows: .venv\Scripts\activate
pip install -r requirements.txt
3. Configure Your MCP Client
Claude Code (~/.claude/settings.json)
{
"mcpServers": {
"seo-audit": {
"command": "/path/to/mcp-gsc/.venv/bin/python",
"args": ["/path/to/mcp-gsc/gsc_server.py"],
"env": {
"GSC_OAUTH_CLIENT_SECRETS_FILE": "/path/to/client_secrets.json"
}
}
}
}
Claude Desktop (claude_desktop_config.json)
Same JSON structure — see Claude Desktop MCP docs for config file location.
4. Optional: CrUX API Key
For Core Web Vitals data, set CRUX_API_KEY in the env block:
"env": {
"GSC_OAUTH_CLIENT_SECRETS_FILE": "/path/to/client_secrets.json",
"CRUX_API_KEY": "your-google-api-key"
}
Get one from Google AI Studio or the Cloud Console with CrUX API enabled.
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
GSC_OAUTH_CLIENT_SECRETS_FILE | OAuth | client_secrets.json | Path to OAuth client secrets |
GSC_CREDENTIALS_PATH | Service account | service_account_credentials.json | Path to service account key |
GSC_SKIP_OAUTH | No | false | Set to true to skip OAuth and use service account only |
GSC_DATA_STATE | No | all | all = fresh data matching GSC dashboard, final = confirmed data (2-3 day lag) |
CRUX_API_KEY | No | — | Google API key for Core Web Vitals (CrUX) |
Example Prompts
"List my GSC properties"
"Show search analytics for cdljobscenter.com last 28 days"
"Find striking distance keywords for my site"
"Detect keyword cannibalization"
"Run a full site audit"
"Check Core Web Vitals for cdljobscenter.com"
"Inspect indexing status of these URLs: /jobs, /companies, /pricing"
"Request indexing for https://mysite.com/new-page"
"Compare search performance this month vs last month"
Tests
71 tests covering all 23 tools with mocked Google API calls:
# Activate venv first
python -m pytest test_gsc_server.py -v
What Changed From the Original
- 23 tools (up from ~10) — added Indexing API, CrUX, SEO analysis, site audit
- 7 bug fixes — sort direction mapping, origin/URL detection, empty rows crash, API key leak, blocking sleep, service caching, stale cache on reauth
- 71-test QA suite — full coverage with mocked API calls
- Security — API keys redacted from error messages
- Performance — Google API service objects cached, async sleep instead of blocking
License
MIT. See LICENSE.
Based on AminForou/mcp-gsc.
相关服务器
bioRxiv
Search and retrieve articles from bioRxiv, the preprint server for biology.
SearXNG MCP Server
A web search server powered by the SearXNG API.
Erick Wendel Contributions
Query Erick Wendel's contributions, including talks, blog posts, and videos, using natural language.
Agntic AI for Research Papers
Search and extract information about research papers from arXiv.
Ferengi Rules of Acquisition
Provides the Ferengi Rules of Acquisition with powerful search and retrieval capabilities.
NRT Search
A near real-time search server for indexing and querying documents, implemented in Java.
EU Regulations MCP
Query 37 EU regulations (DORA, NIS2, GDPR, AI Act, CRA) with full-text search, cross-regulation comparison, and ISO 27001/NIST CSF control mappings. Auto-updates via EUR-Lex monitoring.
FetchSERP
All-in-One SEO & Web Intelligence Toolkit API from FetchSERP.
Google Maps
An MCP server for interacting with the Google Maps API, designed for Google Cloud Run.
Researcher MCP
A research assistant powered by Perplexity AI for intelligent search, documentation retrieval, and code assistance.