Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Talk to me in plain language and gut check where we are at.
.claude/skills/sterlingcrispin-sanitycheck/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 37% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -24% | 0% |
Talk to me in plain language and gut check where we are at.
1) What is your understanding of what we are building? 2) Why are we building it? 3) What is the telos of it? 4) How are we approaching it? 5) What have we done so far? 6) How close are we to being done? 7) Do you have a clearly defined way of knowing you are done? 8) What is your current assessment of our approach, and progress? 9) What do you think we should keep doing, and change, if anything? 10) Any additional feedback or reflections or questions.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 10,348 | 12,813 | +24% | 1 | 1 | 0% | 1,519 | 2,049 | +35% | 0 | 0 | — |
case-02 | fail→pass | 13,862 | 13,088 | -6% | 1 | 1 | 0% | 1,953 | 2,147 | +10% | 0 | 0 | — |
case-03 | fail→pass | 15,402 | 15,584 | +1% | 1 | 1 | 0% | 2,222 | 2,325 | +5% | 0 | 0 | — |
case-04 | pass→fail | 19,299 | 20,956 | +9% | 1 | 1 | 0% | 2,856 | 3,298 | +15% | 0 | 0 | — |
case-05 | pass→fail | 18,184 | 23,026 | +27% | 1 | 1 | 0% | 2,860 | 3,779 | +32% | 0 | 0 | — |
case-06 | pass→fail | 12,597 | 20,986 | +67% | 1 | 1 | 0% | 2,214 | 3,444 | +56% | 0 | 0 | — |
case-07 | fail→pass | 15,141 | 18,066 | +19% | 1 | 1 | 0% | 2,088 | 2,855 | +37% | 0 | 0 | — |
case-08 | fail→pass | 19,160 | 13,241 | -31% | 1 | 1 | 0% | 2,708 | 2,062 | -24% | 0 | 0 | — |
case-09 | fail→pass | 20,700 | 17,614 | -15% | 1 | 1 | 0% | 3,036 | 2,717 | -11% | 0 | 0 | — |
case-10 | fail→pass | 17,483 | 15,093 | -14% | 1 | 1 | 0% | 2,740 | 2,320 | -15% | 0 | 0 | — |
case-11 | fail→pass | 19,403 | 17,752 | -9% | 1 | 1 | 0% | 2,764 | 2,802 | +1% | 0 | 0 | — |
case-12 | fail→pass | 19,927 | 17,289 | -13% | 1 | 1 | 0% | 3,029 | 2,698 | -11% | 0 | 0 | — |
case-13 | fail→pass | 18,911 | 18,021 | -5% | 1 | 1 | 0% | 2,749 | 2,649 | -4% | 0 | 0 | — |
case-14 | fail→pass | 17,172 | 14,388 | -16% | 1 | 1 | 0% | 2,593 | 2,149 | -17% | 0 | 0 | — |
case-15 | fail→pass | 14,510 | 15,990 | +10% | 1 | 1 | 0% | 2,215 | 2,501 | +13% | 0 | 0 | — |
case-16 | pass→pass | 20,903 | 17,925 | -14% | 1 | 1 | 0% | 2,963 | 2,711 | -9% | 0 | 0 | — |
case-17 | fail→fail | 15,844 | 14,522 | -8% | 1 | 1 | 0% | 2,438 | 2,221 | -9% | 0 | 0 | — |
case-18 | fail→pass | 20,421 | 16,425 | -20% | 1 | 1 | 0% | 3,167 | 2,450 | -23% | 0 | 0 | — |
case-19 | fail→pass | 21,371 | 14,929 | -30% | 1 | 1 | 0% | 3,456 | 2,335 | -32% | 0 | 0 | — |
case-20 | fail→pass | 15,490 | 17,345 | +12% | 1 | 1 | 0% | 2,275 | 2,745 | +21% | 0 | 0 | — |
case-21 | fail→pass | 14,031 | 17,755 | +27% | 1 | 1 | 0% | 2,159 | 2,679 | +24% | 0 | 0 | — |
case-22 | fail→pass | 13,244 | 15,583 | +18% | 1 | 1 | 0% | 2,078 | 2,436 | +17% | 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. The headline lift of +64 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
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.