Think inside your AI world.
How do I connect PostHog to Claude?
Two live answers here, not one: PostHog's own free hosted MCP server, and PostHog through Unl, where reads arrive measured against the discipline you ratified, present-tense today.
Direct answer: PostHog runs a free hosted MCP server with roughly eighty tools — trends, funnels, retention, HogQL queries, insights, dashboards, feature flags, experiments with results, error tracking, cohorts, session data, LLM cost. Add it in Claude via Settings → Connectors ('claude mcp add' in Claude Code) and the analytics surface is conversational. Separately, PostHog is live through Unl's own reach lane, and the lane is open to any tool that speaks MCP, so the measured read described on every page of this estate is simply how it works now.
The naked read, at its best
PostHog's server is one of the richest in the ecosystem: ask for the activation funnel, the week's trend, an experiment's current results or a HogQL cut and it arrives structured. If analytics access were the whole problem, this server would end the story.
Where even eighty tools stop
Worked case: you ratified an experiment-stopping rule — no test gets called before two full weeks and a minimum sample, because an early call once shipped a change the next fortnight reversed. PostHog reports significance the moment maths allows; your two-week discipline is a decision about how you use the maths, and it lives with you.
The lane that is already live
Read through Unl, the experiment read arrives inside your rule: nine days in, sample short of your floor — not callable yet, by the discipline you set after the reversal. Same data PostHog serves everyone; what arrives is your judgement applied to it. This page's claim is the estate's most literal: for PostHog, the measured lane is not a pattern, it is the shipped path.
PostHog to Claude is a free hosted server and minutes of setup; PostHog through Unl is live today: every analytics read arrives measured against the analytical discipline you ratified.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. The reach lane is live: one box, paste anything. If it speaks MCP, Unl can reach it. Readings arrive unprompted, the data beside the criterion; Unl is a courier, not a warehouse, and keeps only your keys and the frame.
Read further
Questions people ask
What does the PostHog MCP server include?
Around eighty tools: trends, funnels, retention, HogQL, insights, dashboards, feature flags, experiments with results, error tracking, cohorts, session data and LLM cost tracking — hosted, free, and unusually comprehensive.
What does 'live through Unl' mean concretely?
Unl's reach lane is live and open: reads of your PostHog data arrive inside your ratified criteria now, in the present tense, and unprompted when a turn touches the criterion. The lane is not a provider list; one box, paste anything, and if it speaks MCP, Unl can reach it.
Do I choose between the two connections?
No — they answer different questions. PostHog's own server gives Claude the full analytics toolbox. Through Unl, reads arrive measured against what you decided. Many users run both: the toolbox for exploration, the frame for judgement.
What this is
Think inside your AI world — you stay in command
Unlimitless (Unl to friends) holds what you've settled, reads what your tools are showing, and catches what's changed out in the world — and hands your AI whatever bears on the work, the moment it's needed, without you asking. The right thing, in front of the model, unprompted, with you in command of the call. So you keep moving toward what you set out to build, on top of everything you've already decided.
It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector. Quick to connect, in a couple of steps.
Unlimitless is open now to invited Alpha. Apply for the Beta waitlist to come in ahead of the full launch:
Alpha is invite-only · free at launch.