skill-evaluator

Evaluating skill feedback histories involves aggregating ok, broken, and needs-improvement counts to generate actionable recommendations. The evaluator analyzes these quality signals alongside pending refinement flags and maturity stages to output a concrete promote, refine, or continue decision per skill.

Skill lifecycle management uses feedback evaluation to determine if a skill is ready for promotion or requires refinement. The mechanism reads aggregated quality signals and pending refinement flags from both workspace and installed home skills, producing a concrete recommendation for continued usage.

You can automate skill evaluation across workspace skills and installed home skills by integrating with evaluation workflows like git-skill-manager evaluate. This allows you to automatically read feedback histories and generate promotion, refinement, or continued usage recommendations without manual intervention.

The skill evaluator requires no external dependencies to analyze maturity stages and feedback counts. It operates independently on scripts to read quality signals and generate actionable recommendations for your skills.

Generating actionable recommendations for skill upgrades is best done by running the evaluator to analyze feedback histories. It reads ok, broken, and needs-improvement counts alongside pending refinement flags to produce a concrete promote, refine, or continue decision for all skills.

Skill evaluation recommends continued usage instead of promotion when feedback histories lack sufficient positive quality signals. If the analyzer sees high needs-improvement or broken counts, or pending refinement flags, it determines the skill is not yet mature enough to promote.