Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Compute coverage completeness, redundancy, and gap severity scores from a coverage map.
.claude/skills/yogsoth-ai-coverage-scoring/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 210% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 159% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 300% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 158% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 90% | 0% |
Compute quantitative scores for portfolio coverage completeness, redundancy levels, and gap severity from the candidate-to-niche coverage map.
Spawns a subagent that applies scoring algorithms to the coverage map and produces actionable metrics.
Scoring requires consistent application of metrics across the full coverage map with attention to edge cases (partial coverage, weighted importance). Focused execution ensures no niches are overlooked.
Output must include a numeric coverage score (0-1), redundancy score (0-1), and severity-ranked gap list. All scores must be justified with methodology.
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | spawn-agent | Spawn a customized CC subagent with full MCP tool access. Used by SOPs that declare execution: subagent. |
<!-- END available-tables (generated) -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 11,701 | 42,025 | +259% | 1 | 1 | 0% | 1,839 | 6,414 | +249% | 0 | 0 | — |
case-02 | fail→fail | 14,581 | 28,705 | +97% | 1 | 1 | 0% | 2,326 | 6,409 | +176% | 0 | 0 | — |
case-03 | fail→fail | 17,253 | 9,331 | -46% | 1 | 1 | 0% | 2,684 | 773 | -71% | 0 | 0 | — |
case-04 | fail→pass | 11,153 | 29,100 | +161% | 1 | 1 | 0% | 1,733 | 5,380 | +210% | 0 | 0 | — |
case-05 | fail→pass | 7,996 | 16,251 | +103% | 1 | 1 | 0% | 1,172 | 3,032 | +159% | 0 | 0 | — |
case-06 | fail→pass | 7,488 | 26,905 | +259% | 1 | 1 | 0% | 1,297 | 5,184 | +300% | 0 | 0 | — |
case-12 | pass→pass | 21,749 | 22,151 | +2% | 1 | 1 | 0% | 4,331 | 4,460 | +3% | 0 | 0 | — |
case-07 | fail→pass | 6,483 | 25,091 | +287% | 1 | 1 | 0% | 1,023 | 2,639 | +158% | 0 | 0 | — |
case-08 | fail→fail | 6,266 | 4,716 | -25% | 1 | 1 | 0% | 1,000 | 516 | -48% | 0 | 0 | — |
case-09 | fail→pass | 7,923 | 13,579 | +71% | 1 | 1 | 0% | 1,292 | 2,452 | +90% | 0 | 0 | — |
case-10 | fail→fail | 19,956 | 36,538 | +83% | 1 | 1 | 0% | 3,490 | 6,393 | +83% | 0 | 0 | — |
case-11 | fail→pass | 17,511 | 23,120 | +32% | 1 | 1 | 0% | 2,696 | 4,336 | +61% | 0 | 0 | — |
case-13 | fail→pass | 16,342 | 15,935 | -2% | 1 | 1 | 0% | 2,496 | 3,127 | +25% | 0 | 0 | — |
case-14 | fail→pass | 12,513 | 6,233 | -50% | 1 | 1 | 0% | 1,890 | 1,192 | -37% | 0 | 0 | — |
case-15 | fail→fail | 7,700 | 9,038 | +17% | 1 | 1 | 0% | 1,208 | 734 | -39% | 0 | 0 | — |
case-16 | fail→fail | 4,548 | 18,481 | +306% | 1 | 1 | 0% | 737 | 3,015 | +309% | 0 | 0 | — |
case-17 | fail→fail | 17,644 | 23,398 | +33% | 1 | 1 | 0% | 2,995 | 4,583 | +53% | 0 | 0 | — |
case-18 | fail→pass | 16,773 | 11,045 | -34% | 1 | 1 | 0% | 2,764 | 1,997 | -28% | 0 | 0 | — |
case-19 | fail→pass | 14,594 | 30,059 | +106% | 1 | 1 | 0% | 2,239 | 6,012 | +169% | 0 | 0 | — |
case-20 | pass→pass | 14,502 | 14,038 | -3% | 1 | 1 | 0% | 2,301 | 2,414 | +5% | 0 | 0 | — |
case-21 | pass→pass | 23,276 | 19,008 | -18% | 1 | 1 | 0% | 4,903 | 4,143 | -16% | 0 | 0 | — |
case-22 | pass→pass | 18,569 | 17,087 | -8% | 1 | 1 | 0% | 2,639 | 2,689 | +2% | 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 19 counted toward the lift figure. The other 3 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 +45 percentage points is the difference between those two pass rates over the 19 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.