feat(ms-ai-architect): Sesjon 15 — B2 K10 søsken-scope-ikke-overlapp

- Refaktor: overlap-kjerne flyttet til scripts/kb-eval/lib/sibling-overlap.mjs
  (bryter sirkulær import eval.mjs<->detect); detect re-eksporterer → B1-tester urørt
- K10 = søsken-scope-ikke-overlapp, deterministisk cross-skill: perSkillSiblingOverlap
  + attachSiblingOverlap i eval.mjs. combined = boundaryTension + df-vektet leksikalsk;
  per-skill verdikt = verste søskenpar; terskel 7.0 (naturlig gap 7.42→6.67)
- Empirisk (alle 5): eng+infra FAIL (7.42 mot hverandre), advisor/gov/sec PASS
  → eng↔infra-signal mater B3 merge/saner (ikke blokkering)
- Gated baseline-regen via --write (descriptions urørt → judge K1/K4/K7/K8/K9 merget
  uendret, ikke fabrikkert); rubric K1-K10
- TDD: +9 tester (tests/kb-eval/test-k10-sibling-overlap.test.mjs), kb-eval 31→40
- 0 skriving til skills/. Suiter: validate 239 · kb-update 122 · kb-integrity 192/192
This commit is contained in:
Kjell Tore Guttormsen 2026-06-20 11:59:59 +02:00
commit ba597eb988
5 changed files with 400 additions and 123 deletions

View file

@ -1,6 +1,6 @@
{
"rubric": "K1-K9",
"note": "Deterministic: K2,K3,K5,K6,refCountConsistency. LLM-judge (operator-gated): K1,K4,K7,K8,K9.",
"rubric": "K1-K10",
"note": "Deterministic: K2,K3,K5,K6,refCountConsistency,K10(siblingScopeOverlap). LLM-judge (operator-gated): K1,K4,K7,K8,K9.",
"skills": [
{
"name": "ms-ai-advisor",
@ -55,6 +55,12 @@
"2026",
"v3.0"
]
},
"K10_siblingScopeOverlap": {
"maxCombined": 4.5833,
"worstSibling": "ms-ai-engineering",
"pass": true,
"threshold": 7
}
},
"judgeInputs": {
@ -158,6 +164,12 @@
"preview",
"GA"
]
},
"K10_siblingScopeOverlap": {
"maxCombined": 7.4167,
"worstSibling": "ms-ai-infrastructure",
"pass": false,
"threshold": 7
}
},
"judgeInputs": {
@ -253,6 +265,12 @@
"sample": [
"2024"
]
},
"K10_siblingScopeOverlap": {
"maxCombined": 6.6667,
"worstSibling": "ms-ai-engineering",
"pass": true,
"threshold": 7
}
},
"judgeInputs": {
@ -348,6 +366,12 @@
"preview",
"GA"
]
},
"K10_siblingScopeOverlap": {
"maxCombined": 7.4167,
"worstSibling": "ms-ai-engineering",
"pass": false,
"threshold": 7
}
},
"judgeInputs": {
@ -443,6 +467,12 @@
"sample": [
"2025"
]
},
"K10_siblingScopeOverlap": {
"maxCombined": 6.5,
"worstSibling": "ms-ai-governance",
"pass": true,
"threshold": 7
}
},
"judgeInputs": {