A checkpoint before your coding agent ships.
Skene reads your repository, knows which analytics events should fire and where, and flags removed, renamed, or moved instrumentation on the PR before it merges.
The problem
Event streams are how you make decisions. They break quietly.
Product, growth, and exec all run on event data. Funnels, retention, attribution, pricing experiments. The charts that drive roadmap calls and budget calls come from a stream of posthog.capture and mixpanel.track calls sitting in your repo.
That stream is fragile. A coding agent rewriting a checkout flow does not know which line in the file fires event: "checkout_completed". It changes the function, ships the PR, and your funnel goes quiet. The dashboard keeps drawing, just from worse data. Nobody notices until a number looks wrong three sprints later.
Skene reads every analytics call in your repo.
PostHog capture, Mixpanel track, Segment identify, Amplitude event, Heap track. Indexed by file, line, and payload shape.
Skene parses the codebase the way a linter does. It finds every call into your analytics library, builds an index of where each event fires, and remembers what payload it ships. That index is the baseline every PR gets checked against.
Open GitHub
Drop one capture call. Your dashboard goes quiet for three sprints.
Removed events. Renamed events. Moved events. Altered payloads. Conditional firing that no longer fires.
When a PR changes a file with analytics in it, Skene compares the new state against the index. Missing calls, renamed events, payload shape changes, and conditional firing that no longer fires all show up as a comment on the PR. Caught on review, not on a dashboard that has been wrong since a refactor weeks ago.
Four ways to plug Skene in.
MCP server for your coding agent. GitHub Action for CI. Cloud API for custom runs. One-time repo audit.
Install the MCP server into Cursor or Claude Code and Skene runs before the agent commits. Add the GitHub Action and it runs on every PR. Hit the cloud API directly from any script. Or run a one-time audit of your current instrumentation surface before you adopt anything else.
From install to a clean instrumentation surface.
A baseline of every analytics call you already have, drift caught on the PR that introduced it, and an instrumentation surface that stays consistent as the codebase grows.
See what your last 10 PRs quietly broke.
Install in a minute. The first repo audit runs in about as long as your CI does.
