Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyze implementation gap between requirements and existing codebase
.claude/skills/bilal140202-kiro-validate-gap/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 0% | 0% |
<background_information>
</background_information>
<instructions>
Analyze implementation gap for feature $1 based on approved requirements and existing codebase.
{{KIRO_DIR}}/specs/$1/spec.json for language and metadata{{KIRO_DIR}}/specs/$1/requirements.md for requirementsproduct.md, tech.md, structure.mdrules/gap-analysis.md from this skill's directory for comprehensive analysis frameworkThe following research areas are independent and can be executed in parallel:
If multi-agent is enabled, spawn sub-agents for each area above. Otherwise execute sequentially.
After all parallel research completes, synthesize findings for gap analysis.
Write the gap analysis to disk so it survives session boundaries and can be referenced during design phase.
{{KIRO_DIR}}/specs/$1/research.md---) rather than overwriting previous research</instructions>
Provide output in the language specified in spec.json with:
Format Requirements:
/kiro-spec-requirements $1 first to generate requirements"en) if spec.json doesn't specify languageIf Gap Analysis Complete:
/kiro-spec-design $1 to create technical design document/kiro-spec-design $1 -y to auto-approve requirements and proceed directlyNote: Gap analysis is optional but recommended for brownfield projects to inform design decisions.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-17 | fail→pass | 12,208 | 3,936 | -68% | 1 | 1 | 0% | 1,642 | 1,602 | -2% | 0 | 0 | — |
case-01 | fail→fail | 2,164 | 6,060 | +180% | 1 | 1 | 0% | 334 | 1,426 | +327% | 0 | 0 | — |
case-02 | fail→fail | 8,621 | 5,290 | -39% | 1 | 1 | 0% | 1,385 | 1,351 | -2% | 0 | 0 | — |
case-03 | fail→fail | 5,779 | 5,564 | -4% | 1 | 1 | 0% | 214 | 1,385 | +547% | 0 | 0 | — |
case-04 | fail→pass | 10,853 | 10,653 | -2% | 1 | 1 | 0% | 1,647 | 1,651 | +0% | 0 | 0 | — |
case-05 | fail→fail | 8,466 | 5,930 | -30% | 1 | 1 | 0% | 1,318 | 1,366 | +4% | 0 | 0 | — |
case-06 | fail→fail | 9,348 | 1,945 | -79% | 1 | 1 | 0% | 1,526 | 1,426 | -7% | 0 | 0 | — |
case-07 | fail→fail | 13,008 | 8,425 | -35% | 1 | 1 | 0% | 1,962 | 1,569 | -20% | 0 | 0 | — |
case-08 | pass→pass | 6,091 | 2,312 | -62% | 1 | 1 | 0% | 895 | 1,305 | +46% | 0 | 0 | — |
case-09 | fail→pass | 13,467 | 3,915 | -71% | 1 | 1 | 0% | 1,966 | 1,543 | -22% | 0 | 0 | — |
case-10 | fail→fail | 6,275 | 1,771 | -72% | 1 | 1 | 0% | 858 | 1,337 | +56% | 0 | 0 | — |
case-11 | fail→fail | 12,416 | 7,091 | -43% | 1 | 1 | 0% | 1,996 | 2,109 | +6% | 0 | 0 | — |
case-12 | fail→pass | 7,282 | 2,159 | -70% | 1 | 1 | 0% | 1,152 | 1,402 | +22% | 0 | 0 | — |
case-13 | fail→pass | 9,489 | 2,079 | -78% | 1 | 1 | 0% | 1,402 | 1,404 | +0% | 0 | 0 | — |
case-14 | pass→pass | 8,742 | 3,059 | -65% | 1 | 1 | 0% | 1,353 | 1,476 | +9% | 0 | 0 | — |
case-15 | fail→fail | 4,828 | 2,515 | -48% | 1 | 1 | 0% | 609 | 1,436 | +136% | 0 | 0 | — |
case-16 | fail→pass | 12,109 | 4,609 | -62% | 1 | 1 | 0% | 1,942 | 1,801 | -7% | 0 | 0 | — |
case-18 | pass→pass | 12,353 | 6,312 | -49% | 1 | 1 | 0% | 1,885 | 2,057 | +9% | 0 | 0 | — |
case-19 | fail→pass | 11,930 | 3,664 | -69% | 1 | 1 | 0% | 1,665 | 1,600 | -4% | 0 | 0 | — |
case-20 | fail→fail | 30,958 | 5,668 | -82% | 1 | 1 | 0% | 6,173 | 1,389 | -77% | 0 | 0 | — |
case-21 | fail→fail | 18,233 | 8,532 | -53% | 1 | 1 | 0% | 3,312 | 1,334 | -60% | 0 | 0 | — |
case-22 | fail→fail | 8,362 | 5,951 | -29% | 1 | 1 | 0% | 1,593 | 1,339 | -16% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 14 counted toward the lift figure. The other 8 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +32 percentage points is the difference between those two pass rates over the 14 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.