|
|
92cd93771b
|
feat(ms-ai-architect): Spor D Steg B — skill-quality lifecycle-gate + score-cache (TDD)
B1: gateSkill(report, name) — pure post-mutation verdict (blocked/provisional/
improvements), wired into apply-skill-op.mjs to re-score the affected skill from
fresh disk after create/merge/sanitize-apply. K10 floor enforced immediately;
unjudged → provisional + nudge to re-run the judge pass. retire → skipped.
B2: buildScoreCache(result) — compact, deterministic cache shape + score-skill.mjs
--write [path] → data/skill-score-report.json (always whole-corpus). Gitignored
(derived/regenerable; avoids churn in the public repo). Consumed by Steg C surfacing.
8 new tests (tests/kb-eval/test-skill-score-gate.test.mjs). kb-eval 150 pass,
validate 239 pass. Live score unchanged: advisor 91, eng/gov/infra/sec 96, 0 < 90.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-23 18:26:29 +02:00 |
|
|
|
c1a09062d4
|
feat(ms-ai-architect): Spor D steg 2 — score-skill CLI (--json/--gate/--skill) (TDD)
Byggekloss for både SessionStart-surfacing og lifecycle-gate (D3).
- eval.mjs: ekstraher buildReport() (eksportert, testbar) ut av main() —
samler hele korpus-rapporten (deterministisk + K10 + merget judge-cache).
- lib/skill-score.mjs: formatScoreReport() (ren) — sortert forbedringsrapport
med score, status, floored/provisional/ujudget-tags og ⚑gulv-markering.
- score-skill.mjs (NY CLI): buildReport → scoreReport → render/json/gate.
--skill scorer hele korpuset (K10 trenger alle søsken) men rapporterer/gater
kun den valgte → inkrementell enkelt-skill-scoring for D3 lifecycle-gate.
- docs/development.md: ny «Skill-kvalitetsscore (Spor D)»-seksjon.
Live baseline: advisor 91, governance/security 96 (OK); engineering +
infrastructure 89 (floored av K10-gulv) UNDER mål. Gate exit=1.
Tester: kb-eval 134→142 (+8). validate 239, kb-update 316 uendret grønt.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-23 17:16:44 +02:00 |
|