Blog

Notes on repo health metrics, developer tooling, and shipping indie software. From the team building PulseBoard.

Why we measure 7 repo health signals (and not 17)
More signals sound safer. They aren't. Here's how we narrowed repo health to seven numbers that actually change behavior.
How PulseBoard scores test coverage without running your tests
We don't clone your repo. We don't run pytest. Here's the two-step fallback we use to surface a coverage number you can trust.