Comment by CuriouslyC

Just released a Code Quality leaderboard based on the static analysis research I've been doing. https://codehealth.sibylline.dev/

Now working on comprehensive benchmarks for another tool I built, https://github.com/sibyllinesoft/scribe. Results thus far showing it reducing agent token usage by ~80% in real world tasks, but I need to repeat to get variance. Hopefully I can get a writeup out soon.