Provides tools for querying events, persons, insights, dashboards, feature flags, cohorts, experiments, and actions from a PostHog instance.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@posthog-mcp-server Show me the last 10 pageview events"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
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.
Tool | Description |
| Query events with filters (event name, person, date range, properties) |
| Get a person (user) by distinct_id — properties, creation date, event count |
| Search and list persons with pagination |
| List all dashboards — names, tags, widget counts |
| Get a specific dashboard with its insights and widgets |
| Execute a saved insight (trend, funnel, etc.) and get results |
| List all feature flags — key, active status, rollout percentage |
| Evaluate a flag for a specific user and get variant value |
| List cohorts — name, type (dynamic/static/SQL), person count |
| List A/B tests — name, status, feature flag, dates |
| Get experiment results with variant data and statistical significance |
| List custom event actions/definitions |
| Get project name, ID, settings, and data region |
Related MCP server: Posthog Universal MCP Server
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
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"
}
}
}
}# 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.jsVariable | Required | Default | Description |
| ✅ | — | Your PostHog personal API key ( |
| ❌ |
| PostHog instance URL (use |
| ❌ |
| Default project ID. If empty, must be specified per-call. |
If you run PostHog on your own infrastructure, set POSTHOG_HOST to your instance URL:
POSTHOG_HOST=https://posthog.yourcompany.comOnce 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?"
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
npm install
npm run dev # Build and run
npm run build # Build only
npm start # Run built outputMIT © Nova
Evaluate a feature flag for a specific user. Returns whether the flag is enabled and its variant value.
Inputs: flag_key (required), distinct_id (required).
Execute a saved PostHog insight by ID and return the results. Use list_dashboards to find insight IDs, or pass a raw insight definition.
Inputs: insight_id (required), refresh.
Get a specific PostHog dashboard by ID, including its insights and widgets.
Inputs: dashboard_id (required).
Get detailed results for a specific PostHog experiment, including variants, metrics, and statistical significance.
Inputs: experiment_id (required).
Get a PostHog person (user) by their distinct_id. Returns person properties, creation date, and event count.
Inputs: distinct_id (required).
Get basic information about the PostHog project — name, ID, created date, and available features.
Inputs: no named parameters.
List PostHog event actions (custom event definitions). Shows name, steps, and creation date.
Inputs: limit.
List all PostHog cohorts in the project. Shows name, type, person count, and creation date.
Inputs: limit, search.
List all PostHog dashboards in the project. Returns dashboard names, IDs, and widget counts.
Inputs: limit.
Query events from PostHog. Filter by event name, person, date range, and properties. Returns recent events matching your criteria.
Inputs: after, event, limit, before, +2 more.
List all PostHog experiments (A/B tests) in the project. Shows name, status, and key results.
Inputs: limit.
List all PostHog feature flags in the project. Shows name, key, active status, and rollout percentage.
Inputs: limit, active.
List PostHog persons (users). Search by name or distinct_id, with pagination.
Inputs: limit, offset, search.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 13 tools
Each tool targets a distinct resource-action pair: events, persons, dashboards, insights, feature flags, cohorts, experiments, actions, and project info. The list/get/execute/evaluate verbs clearly differentiate operations on the same resource.
All tool names follow a consistent verb_noun pattern in snake_case (e.g., list_events, get_person, evaluate_feature_flag), with verbs like list, get, execute, and evaluate used uniformly for their respective operations.
13 tools is well-scoped for a PostHog analytics server, covering major data types (events, persons, dashboards, insights, flags, cohorts, experiments, actions) without excessive granularity. Each tool serves a distinct purpose.
The tool set provides comprehensive read-only querying across PostHog's core domains. Minor gaps exist: no direct list_insights endpoint (insights only via dashboards/execute_insight) and no write/update/delete operations, but these are likely intentional for a query-oriented server.
Query your Logspot analytics, companies, and revenue from any AI assistant, and run your actions.
Enable secure connectivity between Sentry issues and debugging data, and LLM clients, using a Model Context Protocol (MCP) server.
Connect AI agents to ProductNow's context engine to search, create, review, and act.
Query your team's drift, vulnerability, and upgrade data from any AI assistant. OAuth 2.1, 51 tools.
