Google SEO MCP

Google SEO MCP

Enables querying Google Search Console and Google Analytics 4 through natural language, with tools for SEO analysis like anomaly detection, cannibalization detection, and opportunity scoring.

Category
Visit Server

README

Google SEO MCP

A Model Context Protocol (MCP) server that transforms how you interact with Google Search Console and Google Analytics 4. Stop exporting CSVs and start asking questions.

📚 View Documentation


License: MIT Tests

Why use this?

❌ The Old, Broken Way

  • Data Silos: Manually checking Google, then GA4.
  • Manual Correlation: Exporting different CSVs and using VLOOKUPs to see if your #1 ranking page actually converts.
  • Switching Fatigue: Logging in and out to manage multiple clients or properties.
  • AI Context Limits: Uploading giant spreadsheets that hit context limits and cause model hallucinations.

✅ The Google SEO MCP Way

  • Platform Synergy: GSC + GA4 in a single context window. Stop exporting, start analyzing.
  • Deterministic Intelligence: The server performs the complex SEO math (cannibalization, Z-score anomalies, striking distance) so your AI agent gets curated insights, not raw data piles.
  • Cross-Platform ROI: Use the Opportunity Matrix to prioritize keywords that have high search visibility (GSC) but poor on-site engagement (GA4).
  • Zero-Config Multi-Account: Connect 20+ accounts. The server automatically resolves the correct credentials for every site URL.

One Server. Two Platforms. Infinite Accounts. Actionable Intelligence.


🎯 Magic Prompts

Copy and paste these into your MCP client (Claude Desktop, etc.) to see the intelligence engine in action:

🔍 The Traffic Detective

"My traffic dropped this week compared to last. Use the anomaly detection and time-series tools to find exactly when the drop started and which pages are responsible."

🎯 The "Striking Distance" Hunter

"Find keywords for https://example.com where I'm ranking in positions 8-15 but have at least 1,000 impressions. These are my best opportunities for a quick traffic boost."

⚔️ The Cannibalization Cleaner

"Check for keyword cannibalization. Are there any queries where two or more of my pages are competing and splitting the traffic? Suggest which one should be the primary authority."

📈 The SEO Opportunity Scoreboard

"Analyze my top 50 keywords for the last 90 days. Rank them by a custom 'Opportunity Score' (Impressions / Position). Give me the top 5 specific pages to focus on."

📊 The Executive Health Check

"Run a full SEO health check for my site. Segment the results by Brand vs. Non-Brand and give me 3 high-impact actions for the upcoming week."

⚡ The Speed vs. Ranking Correlator

"Fetch the top 5 pages by impressions. For these pages, run a PageSpeed audit. Is there any correlation between low performance scores and recently declining positions?"

💰 The ROI Prioritizer (GSC + GA4)

"Run an opportunity_matrix for my top 20 pages. Which high-visibility pages have the lowest engagement or conversion rates? These are my conversion optimization priorities."


🔐 Authentication (Desktop Flow)

Google SEO MCP uses a Secure Desktop Flow. This provides high-security, professional grade authentication for your Google account:

  • Multi-Account Support: Connect multiple Google accounts. The server automatically picks the right one for each site.
  • System Keychain Primary: Tokens are stored in your OS's native credential manager (macOS Keychain, Windows Credential Manager, or Linux Secret Service).
  • AES-256-GCM Hardware-Bound Encryption: Fallback storage is encrypted with AES-256-GCM using a key derived from your unique hardware machine ID. Tokens stolen from your machine cannot be decrypted on another computer.
  • Silent Background Refresh: Tokens auto-refresh silently when they expire.

🚀 Option A: CLI Interactive Setup (Local Desktop)

For local desktop use, run the following command to start the authorization process:

npx google-seo-mcp setup

The CLI will:

  1. Briefly start a secure local server to handle the redirect.
  2. Open your default web browser to the Google Authorization page.
  3. Automatically fetch your email after authorization to label your credentials securely.

🌐 Option B: Express Web OAuth Setup (Centralized/Headless Server)

If you host this MCP server centrally (using Express SSE mode), you can authenticate your Google Account directly via the web browser.

  1. Prerequisite: Set your custom Google OAuth Credentials as environment variables:
    export GOOGLE_CLIENT_ID="your-google-client-id"
    export GOOGLE_CLIENT_SECRET="your-google-client-secret"
    
  2. Register callback: Ensure you register your server's callback URI in the Google Cloud Console: https://mcp.yourcompany.com/search-console/oauth2callback (replace with your custom domain and BASE_PATH).
  3. Trigger Auth: Visit https://mcp.yourcompany.com/search-console/setup in your browser. This will redirect you to Google to log in and authorize the requested scopes.
  4. Complete: Google will redirect you back to the server which automatically saves the OAuth tokens.

Note: You can authorize multiple accounts by visiting this endpoint again and signing in with different Google accounts.

🔑 Step 2 — Logout & Management

To wipe your credentials from both the keychain and the disk:

# Logout of the default account
npx google-seo-mcp logout

# Logout of a specific account
npx google-seo-mcp logout user@gmail.com

🔑 Alternative: Service Account (Advanced)

For server-side environments or automated tasks where interactive login isn't possible, you can use a Google Cloud Service Account.

Setup:

  1. Create Service Account: Go to the Google Cloud Console and create a service account.
  2. Generate Key: Click "Keys" > "Add Key" > "Create new key" (JSON). Download this file.
  3. Share Access: In Google Search Console, add the service account's email address (e.g., account@project.iam.gserviceaccount.com) as a user with at least "Full" or "Restricted" permissions.
  4. Configure: Point the server to your key file:
    export GOOGLE_APPLICATION_CREDENTIALS="/path/to/your/key.json"
    

📊 Google Analytics 4 (GA4)

Connect your GA4 properties to correlate ranking data with user behavior. GA4 can be authorized using either authentication flow:

1. Web OAuth (Recommended for Personal Accounts)

If you used Option B (Express Web OAuth), the scopes for GA4 read-only access were already requested and authorized during that process! Both Search Console and GA4 will immediately work for that account.

2. Service Account (Recommended for Server-to-Server)

If using a service account:

  1. Share Access: Add the service account's email address in GA4 Admin > Property Settings > Property Access Management (grant at least Viewer/Analyst permissions).
  2. Configure Key: Set the key path in your environment variables:
    export GOOGLE_APPLICATION_CREDENTIALS="/path/to/your/key.json"
    

👥 Multi-Account Management

Manage multiple Google accounts from the CLI:

# List all connected accounts
npx google-seo-mcp accounts list

# Remove an account
npx google-seo-mcp accounts remove --account=marketing@company.com

# Add a site boundary to an account
npx google-seo-mcp accounts add-site --account=marketing@company.com --site=example.com

When your AI agent queries a site, the server automatically resolves which account to use. Learn more →


🛡️ Fort Knox Security

This MCP server implements a multi-layered security architecture:

  • Keychain Integration: Primarily uses the macOS Keychain, Windows Credential Manager, or libsecret (Linux) to store tokens.
  • Encrypted Config: Account configuration is stored in ~/.google-seo-mcp-config.enc using AES-256-GCM encryption.
  • Machine Fingerprinting: The encryption key is derived from your unique hardware UUID and OS user. The encrypted file is useless if moved to another machine.
  • Minimalist Storage: Only the refresh_token and expiry_date are stored.
  • Legacy Support: Automatically detects credentials from older versions (tokens files, environment variables).
  • Strict Unix Permissions: Config files are created with mode 600 (read/write only by your user).

Tools Reference

Google Analytics

Tool Description
analytics_query Master tool for raw data. Supports dimensions, filters, aggregationType (byPage/byProperty), dataState (final/all), and type (web/image/news/discover).
analytics_trends Detect trends (rising/falling) for specific queries or pages.
analytics_anomalies Detect statistical anomalies in daily traffic.
analytics_drop_attribution [NEW] Attribute traffic drops to mobile/desktop or correlate with known Google Algorithm Updates.
analytics_time_series [NEW] Advanced time series with rolling averages, seasonality detection, and forecasting.
analytics_compare_periods Compare two date ranges (e.g., WoW, MoM).
seo_brand_vs_nonbrand [NEW] Analyze performance split between Brand vs Non-Brand traffic.

SEO Opportunities (Opinionated)

Tool Description
seo_low_hanging_fruit Find keywords ranking in pos 5-20 with high impressions.
seo_striking_distance [NEW] Find keywords ranking 8-15 (Quickest ROI wins).
seo_low_ctr_opportunities [NEW] Find top ranking queries (pos 1-10) with poor CTR.
seo_cannibalization [Enhanced] Detect pages competing for the same query with traffic conflict.
seo_lost_queries [NEW] Identify queries that lost all traffic in the last 28 days.

SEO Primitives (Atoms for Agents)

These are low-level tools designed to be used by other AI agents to build complex logic.

Tool Description
seo_primitive_ranking_bucket Categorize a position (e.g. "Top 3", "Page 1", "Unranked").
seo_primitive_traffic_delta Calculate absolute and % change between two numbers.
seo_primitive_is_brand Check if a query matches a brand regex.
seo_primitive_is_cannibalized Check if two pages are competing for the same query.

Sites & Sitemaps

Tool Description
sites_list List all verified sites.
sites_add / sites_delete Manage properties.
sites_health_check [NEW] Run a health check on one or all sites. Checks WoW performance, sitemaps, and anomalies.
sitemaps_list / sitemaps_submit Manage sitemaps.

Inspection & Validation

Tool Description
inspection_inspect Google URL Inspection API (Index status, mobile usability).
pagespeed_analyze Lighthouse scores & Core Web Vitals.
schema_validate Validate Structured Data (JSON-LD).

Google Analytics 4 (GA4)

Tool Description
analytics_page_performance Detailed page metrics (sessions, engagement, views).
analytics_traffic_sources Analyze sessions by Channel, Source, and Medium.
analytics_organic_landing_pages Focused metrics for organic traffic landing pages.
analytics_content_performance Analyze content performance by Content Group in GA4.
analytics_conversion_funnel Top converting pages and events.
analytics_user_behavior Device, Country, and Engagement breakdown.
analytics_audience_segments New vs Returning, Age, and OS analysis.
analytics_realtime Live active user data by page and location.
analytics_ecommerce Product and revenue performance.
analytics_pagespeed_correlation Correlate GA4 metrics with PageSpeed scores.

Cross-Platform Intelligence

Tool Description
opportunity_matrix [Flagship] Prioritize SEO tasks by combining signals from GSC and GA4.
page_analysis Joint analysis of ranking (GSC) vs behavior (GA4) for pages.
traffic_health_check Diagnose tracking gaps by comparing GSC clicks to GA4 organic sessions.
brand_analysis Brand vs Non-Brand split across GSC and GA4.

License

MIT Contributing

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured