posthog-mcp-server

posthog-mcp-server

Connects AI assistants to PostHog to query events, persons, insights, dashboards, feature flags, cohorts, and experiments through the Model Context Protocol.

Category
访问服务器

README

PostHog MCP Server

Connect AI assistants to PostHog — query events, persons, insights, dashboards, feature flags, cohorts, and experiments through the Model Context Protocol.

Works with Claude Desktop, Cursor, Windsurf, Cline, Continue, and any MCP-compatible client.

Features

Tool Description
list_events Query events with filters (event name, person, date range, properties)
get_person Get a person (user) by distinct_id — properties, creation date, event count
list_persons Search and list persons with pagination
list_dashboards List all dashboards — names, tags, widget counts
get_dashboard Get a specific dashboard with its insights and widgets
execute_insight Execute a saved insight (trend, funnel, etc.) and get results
list_feature_flags List all feature flags — key, active status, rollout percentage
evaluate_feature_flag Evaluate a flag for a specific user and get variant value
list_cohorts List cohorts — name, type (dynamic/static/SQL), person count
list_experiments List A/B tests — name, status, feature flag, dates
get_experiment Get experiment results with variant data and statistical significance
list_actions List custom event actions/definitions
get_project_info Get project name, ID, settings, and data region

Quick Start

1. Get a PostHog API Key

Go to PostHog → Settings → Personal API Keys and create a key with read access.

Or visit: https://your-instance.posthog.com/settings/user-api-keys

2. Configure your MCP client

Claude Desktop — add to claude_desktop_config.json:

{
  "mcpServers": {
    "posthog": {
      "command": "npx",
      "args": ["-y", "posthog-mcp-server"],
      "env": {
        "POSTHOG_API_KEY": "phx_YOUR_API_KEY_HERE",
        "POSTHOG_HOST": "https://us.i.posthog.com",
        "POSTHOG_PROJECT": "YOUR_PROJECT_ID"
      }
    }
  }
}

Cursor — add to MCP settings (.cursor/mcp.json):

{
  "mcpServers": {
    "posthog": {
      "command": "npx",
      "args": ["-y", "posthog-mcp-server"],
      "env": {
        "POSTHOG_API_KEY": "phx_YOUR_API_KEY_HERE",
        "POSTHOG_HOST": "https://us.i.posthog.com",
        "POSTHOG_PROJECT": "YOUR_PROJECT_ID"
      }
    }
  }
}

3. Run manually (for testing)

# Clone and build
git clone https://github.com/friendlygeorge/posthog-mcp-server.git
cd posthog-mcp-server
npm install
npm run build

# Run
POSTHOG_API_KEY=phx_... POSTHOG_PROJECT=12345 node dist/index.js

Environment Variables

Variable Required Default Description
POSTHOG_API_KEY Your PostHog personal API key (phx_...)
POSTHOG_HOST https://us.i.posthog.com PostHog instance URL (use https://eu.i.posthog.com for EU)
POSTHOG_PROJECT "" Default project ID. If empty, must be specified per-call.

Self-hosted PostHog

If you run PostHog on your own infrastructure, set POSTHOG_HOST to your instance URL:

POSTHOG_HOST=https://posthog.yourcompany.com

Example Queries

Once configured, ask your AI assistant:

  • "Show me the last 20 pageview events"
  • "Who are the most active users this week?"
  • "What's the status of our signup conversion experiment?"
  • "Is the new-checkout-flow feature flag enabled for user abc123?"
  • "Show me all active dashboards"
  • "Execute insight 42 and show me the trend"
  • "List all feature flags that are currently active"
  • "How many people are in the 'Power Users' cohort?"

API Reference

This server wraps the PostHog REST API (v1). Full documentation:

  • PostHog API docs: https://posthog.com/docs/api
  • MCP protocol: https://modelcontextprotocol.io
  • PostHog: https://posthog.com

Development

npm install
npm run dev     # Build and run
npm run build   # Build only
npm start       # Run built output

License

MIT © Nova

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选