From IonBin New tools and workflows every week. See what’s new
Ask product analytics questions from your editor
Business data & reporting · Dev, web & IT · checked 2026-10-16
In Cursor or Claude Code, ask PostHog questions in plain language through its MCP server ("which pages drive signups this month?") and get answers next to your code.
Independent review. IonBin has no partnership with any tool on this page and was not paid for this page. Our evaluation is our own, based on public documentation and our research. Product names are trademarks of their owners and are used for identification only. How disclosure works.
Goal
Answer product questions without building a dashboard first.
Explains the result and suggests the next question
AI platform
Confirmed
3
You
Sanity-check surprising numbers in PostHog before acting
Human step
Confirmed
Confirmed means the vendor documents it (linked). It doesn't mean IonBin has tested it. AI-platform and human steps need no source.
Human approval
You decide what to change based on the answer.
Watch out for
Keep the human approval step: You decide what to change based on the answer.
Starter prompt
Goal: Answer product questions without building a dashboard first. Tools involved: PostHog. Your part: runs the query and returns results; explains the result and suggests the next question. Work step by step, show me each draft or proposed change, and wait for my approval before anything is sent, published, signed, filed or changed in a live system.
Assembled from this page’s goal and steps. Paste it into your AI platform and adjust it to your situation.
Computed from the MCP and API steps only: "Confirmed" needs the tool's docs to name the platform and the platform to support the connection. Browser, upload and human steps work from any platform.
What we checked
PostHog’s MCP server is a free, hosted endpoint that works with Claude Code, Cursor, Codex, VS Code and other MCP clients. It can run analytics queries and SQL, and it covers LLM analytics too: LLM traces, costs per model, and evaluation datasets. Tools that use LLMs internally may bill as PostHog AI spend, and they only appear when AI data processing is on. Sanity-check surprising numbers in PostHog before you act on them.
Set it up
Add PostHog’s MCP server to Cursor, Claude Code or another client, and sign in to your PostHog project.
Ask questions in plain language, such as which pages drive signups this month or where users drop out of onboarding.
Ask the AI to show the query it ran (HogQL or the insight) alongside the answer, so you can open the same view in PostHog.
PostHog also offers a VS Code extension with inline analytics and flag management, if you’d rather not chat.
Sources
PostHog MCP · PostHog · official · accessed 2026-10-16