caveman-commit (JuliusBrussee/caveman) vs video (smixs/visual-skills)
caveman-commit (JuliusBrussee/caveman)
26.0 / 30
SituationalSafe
caveman-commit: writes Conventional Commits messages with a breaking-change footer (control 4.0, skill 5.0, one-liner 4.95) for about the same tokens; control writes a plain sentence, and a one-sentence prompt of the format does as well
- SOTAPass
video (smixs/visual-skills)
20.8 / 30
SituationalSafe
smixs video skill: richer, more directed prompts, but keyword checks score it lower and it costs 2.8x the tokens; no video was generated
- SOTAPass