Skip to main content
Glama
Sign In
google-search-console-mcp-claude-code

google-search-console-mcp-claude-code

by mario-hernandez
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python

⚠️ This repo has been superseded

This MCP has been unified with the GA4 MCP into a single comprehensive Google SEO suite — including five cross-platform tools that connect Search Console with Analytics 4 (GSC↔GA4 journey, opportunity matrix, traffic health check, revenue attribution, full landing page diagnosis).

👉 New repo: google-seo-mcp-claude-code

The new MCP includes everything this one did (with prefixed gsc_* tool names) plus the five cross-platform tools that are only possible with unified auth.

Related MCP server: google-search-console-mcp-python

Migration

pipx uninstall gsc-seo-mcp
pipx install git+https://github.com/mario-hernandez/google-seo-mcp-claude-code

Update your Claude config from gsc-seo-mcp to google-seo-mcp.

Tool names are prefixed in the new MCP:

  • quick_wins → gsc_quick_wins

  • traffic_drops → gsc_traffic_drops

  • content_decay → gsc_content_decay

  • (etc.)

This repo is kept for historical reference only and is no longer maintained.

Available Tools

14 tools
alertsA

Detects regressions in last N days vs prior N days.

Emits per-entity (query+page) alerts: position_drop, ctr_collapse, click_drop, disappeared. Severity: critical (≥2× threshold) > warning. Dedup keeps highest severity per entity.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), days, severity_threshold.

cannibalizationC

Queries where >=2 pages on the same site are competing in search results.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), days, min_impressions, top_n.

content_decayB

Pages with monotonic decline across 3 consecutive 30-day windows.

Filters noise: only pages where clicks_p3 > clicks_p2 > clicks_p1 AND clicks_p3 >= threshold. Indicates real content decay, not a single-week dip.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), top_n, min_clicks_p3.

ctr_opportunitiesC

Pages whose CTR is significantly below the expected CTR for their position.

CTR gap = expected_ctr(pos) - actual_ctr. Tipo "title/meta needs improvement".

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), days, min_impressions, top_n.

get_capabilitiesA

List all tools exposed by this MCP and current auth status. Call FIRST.

Returns the tool catalog grouped by category, plus a quick check of whether credentials are reachable.

Inputs: no named parameters.

inspect_urlA

Inspect a single URL against Google's index.

Args: site_url: Property URL (e.g. https://example.com/ or sc-domain:example.com). page_url: Full URL to inspect (must belong to site_url). language: BCP-47 lang for human-readable strings.

Returns indexing status, last-crawl, canonical, mobile usability, rich-results.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), page_url (required), language.

list_sitemapsB

List sitemaps submitted for a property, with errors/warnings/last-submitted info.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required).

list_sitesA

List all Search Console properties the authenticated user has access to.

Returns each site's URL, permission level, and whether it's a domain or URL property.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: no named parameters.

quick_winsB

Queries currently ranking position 4-15 with highest "lift opportunity".

Opportunity score = impressions × (CTR_at_pos3 - CTR_actual). Highlights queries where a small ranking improvement could yield disproportionate traffic gains.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), days, min_impressions, top_n.

reauthenticateA

Force re-authentication on the next API call.

Useful when ADC credentials have changed or OAuth token has expired and cached state is stale. Does not delete files; just resets in-process clients.

Inputs: no named parameters.

search_analyticsA

Custom Search Analytics query.

Args: site_url: Property URL. start_date / end_date: ISO YYYY-MM-DD. dimensions: e.g. ["query"], ["page"], ["query","page"], ["country"], ["device"]. row_limit: Max rows (server-side cap 25000). search_type: "web" | "image" | "video" | "news".

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), start_date (required), end_date (required), dimensions, +2 more.

site_snapshotB

Aggregated totals (clicks/impressions/CTR/position) for the last N days vs prior period.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), days.

submit_sitemapA

Submit a sitemap. Requires GSC_ALLOW_DESTRUCTIVE=true.

Args: site_url: Property URL. feedpath: Full sitemap URL (e.g. https://example.com/sitemap.xml).

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), feedpath (required).

traffic_dropsA

Pages that lost traffic, classified into Ranking-loss / CTR-collapse / Demand-decline.

  • Ranking loss: avg position got worse by >2 positions.

  • CTR collapse: position stable but CTR < 70% of prior CTR.

  • Demand decline: rankings & CTR roughly stable but impressions dropped >30%.

IMPORTANT: Use ONLY the data returned by this tool. Do not speculate about figures, do not extrapolate beyond the time range queried, and cite site_url + date_range when reporting numbers to the user.

Inputs: site_url (required), days, top_n, min_clicks_prior.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.7/5.0

    Scored across 14 tools

    Disambiguation5/5

    Each tool targets a distinct concept: alerts for regressions, cannibalization for competing pages, content_decay for declining pages, etc. No two tools have overlapping purposes; descriptions clearly differentiate them.

    Naming Consistency4/5

    All tool names use lowercase with underscores. While some are verb_noun (get_capabilities, inspect_url) and others are noun phrases (alerts, cannibalization, content_decay), the pattern is consistent and readable, with minimal deviation.

    Tool Count5/5

    14 tools is well-scoped for a Search Console server. It covers analytics, diagnostics, management, and specialized insights without being overwhelming or too thin.

    Completeness4/5

    The tool surface covers core Search Console operations: analytics (search_analytics, site_snapshot), diagnostics (inspect_url, list_sitemaps), and specialized insights (alerts, cannibalization, etc.). Minor gaps exist (e.g., no tool for managing users or viewing all errors), but the set is comprehensive for its focus.

    Maintenance

    ActivityInactive
    ResponsivenessOutdated data

    Resources

    Related MCP Connectors

    Related MCP Servers

    • flin-google-search-console-mcp

      flin-agency
      A
      license
      B
      quality
      B
      maintenance
      Read-only MCP server for Google Search Console, with browser-based OAuth flow and local token storage, enabling querying search analytics, site lists, and URL inspection.
      9
      516 PyPI
      MIT

    • google-search-console-mcp

      locomotive-agency
      A
      license
      Not graded
      quality
      C
      maintenance
      MCP server for Google Search Console API that enables querying search analytics, managing sites, inspecting URLs, and supporting domain delegation via service accounts.
      34 PyPI
      1
      MIT

    • SearchConsole.ai

      AndreyAzimov
      A
      license
      Not graded
      quality
      C
      maintenance
      Read-only MCP server for Google Search Console data, enabling search analytics, URL inspection, indexing diagnostics, and sitemap management through MCP clients.
      19 npm
      1
      MIT

    • google-search-console-mcp

      russjeffery
      A
      license
      A
      quality
      D
      maintenance
      MCP server for Google Search Console API, enabling search performance analysis, URL inspection, sitemap management, and property listing. Supports local, self-hosted, and Cloudflare Workers deployments.
      11
      MIT