evolve-skill
Benchmark-gated skill evolution — consume bigpowers-benchmark report, propose plan-work change, edit skill via craft-skill, re-run benchmark, record ADR. Use when a skill underperforms on benchmark or stocktake finds systemic gap.
danielvm-git · 已有静态检查记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add danielvm-git/bigpowers --skill evolve-skill技能说明
name: evolve-skill model: opus description: "Benchmark-gated skill evolution — consume bigpowers-benchmark report, propose plan-work change, edit skill via craft-skill, re-run benchmark, record ADR. Use when a skill underperforms on benchmark or stocktake finds systemic gap."
story: e23s03
story: e09s01
story: e09s05
Evolve Skill
HARD GATE — No skill change ships without benchmark score ≥ pre-change baseline. Learning is measured and versioned — never implicit.
Loop
-
Regression gate — Run
bash scripts/run-verification-gates.shto catch mechanical regressions (compliance, sync pipeline, size budget) before spending time on benchmark evals. If golden suite fails, fix regressions first — they are pre-requisites for any capability improvement. -
Establish baseline — Run
run-benchmark <skill> --baseline. If no definition exists atspecs/benchmarks/<skill>.yaml, create one followingspecs/benchmarks/SCHEMA.mdfirst. Save report path instate.yaml. Ifspecs/benchmarks/reports/BASELINE-<skill>.yamlalready exists, skip this step. -
Identify gap — Read the baseline report (
specs/benchmarks/reports/BASELINE-<skill>.yaml). Find scenarios withresult: FAILor lowpass_at_k. This is the measurable gap. -
plan-work— Write a minimal change proposal targeting the failing scenarios. Include verify commands. -
Edit via
craft-skill/ direct SKILL.md edit; runbash scripts/sync-skills.sh. -
Re-run benchmark —
run-benchmark <skill>. Compare newpass_at_kagainst baseline.- IMPROVED or STABLE → advance to step 6.
- REGRESSION (
new pass_at_k < baseline) → revert the change and loop back to step 3.
-
Record decision — Write
specs/adr/NNNN-evolve-<skill>.mdwith before/afterpass_at_kscores. Updatesession-state.
Verify
→ verify: test -d specs/benchmarks && test -f skills/run-benchmark/SKILL.md
See REFERENCE.md for ADR template.
Evolve Skill — ADR snippet
## ADR-XXXX: Evolve <skill-name>
**Status:** Accepted
**Benchmark:** before X% / after Y%
**Change:** one-sentence summary
**Evidence:** path/to/benchmark-report.md
Benchmark repo: bigpowers-benchmark — resolve the local clone on this machine; clone it from wherever your fork lives if absent. Never assume a hard-coded absolute path or a specific owner.