A meter, not a leaderboard.

Nostr's promise is that one identity, one set of posts and one social graph work in any app. That is only true if apps implement the same specs the same way, and until now nobody measured whether they do. Nostrometer is that measurement: every app that announces itself, ranked by real usage, rated NIP-by-NIP with ratings that anyone can publish and anyone can check.

The numbers hold up because nothing in them is private. Every rating is a signed public event. Every tier is a median. Every claim from source code says which file the marker was found in. Every Borked or Isolated finding goes to the app's tracker as an issue. Rerun the crawls against the same relays and you get the same matrix.

It is built and maintained by Derek Ross. The crawls are read-only and every event is signature-verified before it counts. The tooling never signs or publishes anything; ratings are published by people, from NostrHub or any Nostr client.

The data is shared 1:1 with NostrHub. The same kind 31986 events feed both. There is no private score.

The name is the idea. A meter reads what is there. If a cell says Incomplete, the fix is in the app, and the rating changes when the app does. See the methodology for exactly how a number gets on the board.