← Back to directory

PostHog

Official

Open-source product analytics platform PostHog, steerable from Claude Code, Cursor, and other MCP-compatible agents.

GitHub source repository ↗

★ 37.5k Stars Category · Other Very popular

41FMRS · D

Reliability

4/20

Security and permissions

8/20

Maintenance

17/20

Documentation

7/20

Setup experience

5/20

PostHog is a broad, open-source product analytics platform whose README mentions MCP connectivity into Claude Code, Cursor, and other MCP-compatible agents, but the source material does not provide a specific MCP tool list, transport details, or installation config, so those fields are left without fabricated evidence.

Read the FMRS scoring method →

PostHog is an open-source platform for building 'self-driving products,' combining product analytics, web analytics, session replay, feature flags, experiments, error tracking, logs, surveys, data warehouse/pipelines (CDP), and AI observability. Per the repository README, in addition to Slack, the web app, and PostHog Desktop, users can connect PostHog via MCP (posthog.com/mcp) to bring it into Claude Code, Cursor, or any MCP-compatible agent. The README does not enumerate specific MCP tools or provide a client configuration example, so this profile reflects only what is explicitly stated in the source material.

Tools

The tool list has not been reviewed yet.

Setup

The README does not give explicit MCP installation commands or a client config JSON. Based on what is documented: 1) sign up for PostHog Cloud (US or EU); 2) for self-hosting, run the repo's one-line hobby deploy script (bin/deploy-hobby) on Linux with Docker (4GB memory recommended); 3) follow the official instructions at posthog.com/mcp to connect the MCP into Claude Code, Cursor, or another MCP-compatible client. No further detail on MCP-specific setup (auth, keys, config file) is present in the source material.

Fit and risk

Best for

Teams already using PostHog for product analytics, feature flags, or experiments who want to control it from an AI agent
Developers who want product signals (errors, rage clicks, etc.) surfaced directly inside Claude Code, Cursor, or similar agent environments

Not for

Users who only need a narrow, single-purpose MCP tool (e.g., pure database or file-system access) — PostHog's surface area is broad
Users requiring a strictly FOSS codebase — the `ee` directory in this repo is under a separate, non-MIT license
Self-hosted deployments needing vendor support — the README explicitly states no customer support or guarantees are provided for open-source deployments

Required permissions

Access to data within a PostHog project/account, including events, session recordings, feature flags, experiments, error tracking, surveys, and data warehouse records
Write access to modify project configuration if using the agent to create or change feature flags/experiments

Risks and side effects

PostHog aggregates substantial user behavioral and product data; agent access to it carries exposure risk for sensitive information (user identifiers, session content)
Repository licensing is mixed (MIT for the main repo, separate proprietary license for the `ee` directory) — confirm which features are covered before relying on them in production
Self-hosted open-source deployments are explicitly unsupported by the vendor and are recommended only up to roughly 100k events/month before needing to migrate to Cloud

Troubleshooting

For self-hosting issues, consult the official self-hosting and troubleshooting docs referenced in the README
If the one-line hobby deploy script (`bin/deploy-hobby`) fails, verify Docker is installed and the recommended 4GB memory is available
For MCP connection issues with Claude Code/Cursor, refer to the official instructions at posthog.com/mcp — the repository README does not include MCP-specific troubleshooting steps

Use cases

Querying product analytics, web analytics, or session replay data through an agent

Having an agent turn signals like errors or failed queries into diagnostic reports or fix suggestions ('self-driving mode')

Creating or adjusting feature flags and experiments via an agent

Reviewing logs or data warehouse records from within a coding/agent workflow to help diagnose issues

Supported clients

Supported clients have not been confirmed yet.

Related MCP servers

Compare with other MCP servers