caveman-review (JuliusBrussee/caveman) vs mcp-builder (anthropics/skills)
caveman-review (JuliusBrussee/caveman)
25.1 / 30
SituationalSafe
caveman-review: shifts code reviews toward short per-finding lines (control 3.3, skill 4.2, one-liner 4.8) at the same tokens, but the skill's own L<line> line format was followed in only some runs; a one-sentence prompt of the format scores higher
- SOTAPass
mcp-builder (anthropics/skills)
21.6 / 30
SituationalSafe
mcp-builder: lifts MCP servers from 3.6 to 4.6/5 by adding naming, annotations and async, but a one-sentence checklist scores 4.9 for 68% of the tokens
- SOTAPass