Cursor rule

1---
2description: "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."
3alwaysApply: false
4---
5 
6# story: e23s03
7 
8# story: e09s01
9# story: e09s05
10 
11# Evolve Skill
12 
13> **HARD GATE** — No skill change ships without benchmark score ≥ pre-change baseline. Learning is measured and versioned — never implicit.
14 
15## Loop
16 
171. **Regression gate** — Run `bash scripts/run-verification-gates.sh` to 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.
182. **Establish baseline** — Run `run-benchmark <skill> --baseline`. If no definition exists at `specs/benchmarks/<skill>.yaml`, create one following `specs/benchmarks/SCHEMA.md` first. Save report path in `state.yaml`. If `specs/benchmarks/reports/BASELINE-<skill>.yaml` already exists, skip this step.
19 
203. **Identify gap** — Read the baseline report (`specs/benchmarks/reports/BASELINE-<skill>.yaml`). Find scenarios with `result: FAIL` or low `pass_at_k`. This is the measurable gap.
21 
224. **`plan-work`** — Write a minimal change proposal targeting the failing scenarios. Include verify commands.
23 
245. **Edit** via `craft-skill` / direct SKILL.md edit; run `bash scripts/sync-skills.sh`.
25 
266. **Re-run benchmark** — `run-benchmark <skill>`. Compare new `pass_at_k` against baseline.
27   - **IMPROVED or STABLE** → advance to step 6.
28   - **REGRESSION** (`new pass_at_k < baseline`) → revert the change and loop back to step 3.
29 
307. **Record decision** — Write `specs/adr/NNNN-evolve-<skill>.md` with before/after `pass_at_k` scores. Update `session-state`.
31 
32## Verify
33 
34→ verify: `test -d specs/benchmarks && test -f skills/run-benchmark/SKILL.md`
35 
36See [REFERENCE.md](REFERENCE.md) for ADR template.
37 
38 
39<!-- story: e31s07 -->
40 
41---
42 
43# Evolve Skill — ADR snippet
44 
45```markdown
46## ADR-XXXX: Evolve &lt;skill-name&gt;
47 
48**Status:** Accepted
49**Benchmark:** before X% / after Y%
50**Change:** one-sentence summary
51**Evidence:** path/to/benchmark-report.md
52```
53 
54Benchmark 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.
55