ga4-mcp
An MCP server integrating Google Analytics 4, Search Console, and Indexing API, enabling AI agents to run reports, inspect URLs, manage properties, and request indexing.
README
ga4-mcp
GA4 + Google Search Console MCP server. Works with Claude, Copilot, Cursor, Codex, Gemini, and any MCP-compatible AI agent.
16 tools for Google Analytics 4, Search Console, and the Indexing API — run reports, inspect URLs, submit sitemaps, manage properties, and request indexing, all from your AI agent.
Quick Start
npx -y ga4-mcp
Or install globally for faster startup:
npm install -g ga4-mcp
ga4-mcp
Prerequisites
Authenticate with Google (includes all required scopes):
gcloud auth application-default login \
--scopes="https://www.googleapis.com/auth/analytics.readonly,https://www.googleapis.com/auth/analytics.edit,https://www.googleapis.com/auth/webmasters,https://www.googleapis.com/auth/indexing"
Or set GOOGLE_APPLICATION_CREDENTIALS to a service account JSON key file.
Agent Configuration
Claude Code
{
"mcpServers": {
"ga4-analytics": {
"command": "npx",
"args": ["-y", "ga4-mcp"]
}
}
}
Cursor / Windsurf / Continue
Add to .cursor/mcp.json or equivalent:
{
"mcpServers": {
"ga4-analytics": {
"command": "npx",
"args": ["-y", "ga4-mcp"]
}
}
}
Load only specific tool groups
{
"mcpServers": {
"ga4-analytics": {
"command": "npx",
"args": ["-y", "ga4-mcp", "--tools", "ga4,gsc"]
}
}
}
CLI Options
ga4-mcp [options]
--tools <groups> Tool groups to enable: ga4, gsc, indexing, admin, all (default: all)
"ga4" = read-only analytics. "admin" = write/delete mega-tool.
--transport <type> Transport: stdio (default) or http (Streamable HTTP)
--port <number> HTTP port (default: 3000)
--version Show version
--help Show help
Environment variables: GA4_MCP_TOOLS, GA4_MCP_TRANSPORT, GA4_MCP_PORT
Tools (16 total)
Utility (1 tool)
| Tool | Description |
|---|---|
ping |
Health check — returns pong (always registered) |
GA4 Analytics (6 tools) — group: ga4
| Tool | Description |
|---|---|
get_account_summaries |
List all GA4 accounts and properties |
get_property_details |
Property details (name, timezone, currency) |
run_report |
Run GA4 Data API reports |
run_realtime_report |
Run GA4 realtime reports |
get_custom_dimensions_and_metrics |
Custom dimensions and metrics |
list_google_ads_links |
Google Ads links for a property |
GA4 Admin (1 mega-tool, 14 actions) — group: admin
| Tool | Description |
|---|---|
ga4_admin |
Create/update/delete properties, streams, key events, audiences |
Google Search Console (6 tools) — group: gsc
| Tool | Description |
|---|---|
gsc_list_sites |
List all GSC properties |
gsc_search_analytics |
Query search analytics (impressions, clicks, CTR, position) |
gsc_inspect_url |
URL inspection (index status, crawl info, mobile usability) |
gsc_list_sitemaps |
List submitted sitemaps |
gsc_add_site |
Add a site to GSC |
gsc_submit_sitemap |
Submit a sitemap |
Indexing API (2 tools) — group: indexing
| Tool | Description |
|---|---|
indexing_notify |
Notify Google to crawl/remove a URL |
indexing_status |
Check notification status for a URL |
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。