Think inside your AI world.
PostHog MCP server: what it is and what it reads
Free, hosted, and unusually comprehensive — around eighty tools — plus the lane every tool on this site shares: through Unl its reads arrive measured against what you decided.
The PostHog MCP server is PostHog's free hosted Model Context Protocol server, carrying roughly eighty tools: trends; funnels and retention; HogQL and the insights layer; dashboards; feature flags; experiments, results included; error tracking; cohorts and session data; LLM cost tracking. Any MCP client connects. And separately from PostHog's own server: measured reads of PostHog data through Unl are live, shipped behaviour today, and the lane is open to any tool that speaks MCP.
The most generous server in the stack
Free and hosted removes both objections at once, and the tool count covers the practice end to end — from raw HogQL to experiment arms to the LLM cost ledger. If every server were built like this, the access story would simply be over.
Two lanes, one tool, both live
PostHog's own server gives any assistant the analytics toolbox. Unl's reach lane, live today, delivers the measured read: the same figures arriving inside the analytical discipline you ratified, stopping rules and sample floors and metric freezes applied with their reasons. This estate describes that pattern for many tools; for PostHog it is not a pattern, it is the shipped path, present tense.
The PostHog MCP server hands any assistant eighty tools of analytics access, free; through Unl the same data arrives already judged by the discipline you set.
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 cost?
Nothing — it is free and hosted by PostHog, with roughly eighty tools across trends, funnels, retention, HogQL, insights, dashboards, flags, experiments, error tracking, cohorts, sessions and LLM cost.
What does 'live through Unl' mean?
Unl's reach lane is live in production and PostHog is connected through it now — measured reads are current behaviour, not roadmap. The lane is open: one box, paste anything, and if it speaks MCP, Unl can reach it.
Where do I find connection steps?
The connect pages linked below give the per-assistant steps; this page is the server spec both share.
What if the data and your discipline disagree?
Then you have the useful version of the problem: a number and a standard, side by side, with the standard visible. Most analytics arguments are really about an unstated standard. Reading against one you set in advance turns the argument into a decision about whether the standard was right.
What this is
Think inside your AI world — you stay in command
Save the thoughts, decisions and targets worth keeping, each with its reasoning, carried into every AI session the moment they matter. A new unit of exchange between you and your AI: the Settled Why with standing that travels. Unprompted.
Your whole AI world. What you decided at the epicentre. It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector — quick to connect, in a couple of steps.
MCP native·Human settled·Model agnostic·Your data
Measured Context
Connect Unl to bring the right information into the moment.
Your sources, read against the criteria you set.
The full product, open. Free at launch.