skill
ocupação
descrição
atualizado
compare
Desenvolvedores de software
Compare two benchmark runs side-by-side with narrative analysis of struggles, cost, and gate friction.
2026-03-25
changelog
Desenvolvedores de software
Record a 5W1H changelog entry for the latest framework change into DESIGN_EVOLUTION.md. Run after committing changes to capture who/what/when/where/why/how.
2026-03-25
compliance
Analistas de garantia de qualidade de software e testadores
Analyze how well an agent followed its CLAUDE.md strategy rules during a benchmark run. Use when analyzing agent behavior, rule compliance, or debugging why an agent failed.
2026-03-20