Who on GitHub is growing fast, who is slowing down, and where you would actually use it.
58 repos from GitHub Trending and GitHub Search, with real daily star counts from GH Archive, snapshots kept in Turso, and Jev reading each one for its category, the person it serves and the place you would use it. Tap any repo to dig in.
Momentum radar
Where the new stars went this week
Sum of new stars in the last 7 days by Jev category (Jev's lean is used when it was unsure).
Every repo on the radar
| Repo | Category | Stars | Last day | 7 days | 30 days | vs prior | Organic | Jev signal |
|---|
Click a column to sort, a row to open the analysis. Held out: flutter/flutter, NationalSecurityAgency/ghidra (GH Archive history stopped updating for them, so their growth cannot be measured here).
How this is measured
Stars per day come from GH Archive WatchEvents through the OSS Insight API, not from GitHub's stargazer list (GitHub no longer serves that list to third parties). GH Archive misses stars from before a repo went public, so totals use GitHub's own count and the daily numbers use GH Archive.
The quadrant puts new stars per day over the last 7 days on the x axis and that pace divided by the pace of the 3 weeks before on the y axis. The vertical line is the radar median (503 a day); the horizontal line is "same speed as before".
Check first: a repo is kept out of the top 10s and headlines (but still shown) when it is over 60 days old and got 80%+ of all its stars in the last 7 days, when its daily pace is more than 3.5 robust deviations above the rest of the radar, or when 3+ repos from the same owner each got 80%+ of their stars in the same week. These patterns fit launch campaigns and bought stars alike, so they are a reason to look closer, not a verdict. This run: morluto/rea, storytold/photocraft, boykopovar/AnyPS5, storytold/artcraft, storytold/pdfcraft, storytold/wordcraft, storytold/filmcraft, storytold/lightcraft, storytold/vectorcraft.
Jev (jev-1.13.0) answered five typed questions per repo from its GitHub facts and growth numbers: category, who it is for, where you would use it, whether adoption looks organic, and how worth studying it is this week. Any pick under 80% confidence is shown as unsure.
Storage: Turso libSQL database tutorhero-stars: 154 snapshot rows, 4569 daily rows. Each run adds a snapshot so day-over-day changes build up over time.