Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Verify a production fix by linking the changed code and release to Sentry issue state and post-release event trends without resolving issues prematurely.
.claude/skills/devin-axis-sentry-fix-verification/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 85% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 38% | 0% |
Keep local tests, merged code, deployed release, and observed production behavior as separate milestones.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 27,586 | 16,315 | -41% | 1 | 1 | 0% | 2,266 | 612 | -73% | 0 | 0 | — |
case-12 | fail→pass | 12,792 | 9,212 | -28% | 1 | 1 | 0% | 1,338 | 901 | -33% | 0 | 0 | — |
case-02 | fail→fail | 15,835 | 16,032 | +1% | 1 | 1 | 0% | 1,785 | 1,291 | -28% | 0 | 0 | — |
case-03 | fail→fail | 17,661 | 15,836 | -10% | 1 | 1 | 0% | 2,007 | 542 | -73% | 0 | 0 | — |
case-04 | fail→pass | 11,820 | 13,987 | +18% | 1 | 1 | 0% | 1,174 | 1,662 | +42% | 0 | 0 | — |
case-05 | fail→pass | 12,458 | 17,152 | +38% | 1 | 1 | 0% | 1,232 | 2,277 | +85% | 0 | 0 | — |
case-06 | fail→pass | 7,222 | 12,766 | +77% | 1 | 1 | 0% | 1,135 | 1,548 | +36% | 0 | 0 | — |
case-07 | fail→pass | 15,541 | 12,871 | -17% | 1 | 1 | 0% | 997 | 1,373 | +38% | 0 | 0 | — |
case-08 | pass→pass | 16,693 | 12,966 | -22% | 1 | 1 | 0% | 1,997 | 1,473 | -26% | 0 | 0 | — |
case-09 | pass→pass | 12,997 | 9,319 | -28% | 1 | 1 | 0% | 1,390 | 938 | -33% | 0 | 0 | — |
case-10 | fail→pass | 10,366 | 15,949 | +54% | 1 | 1 | 0% | 801 | 1,971 | +146% | 0 | 0 | — |
case-11 | pass→pass | 12,499 | 9,220 | -26% | 1 | 1 | 0% | 1,135 | 785 | -31% | 0 | 0 | — |
case-13 | fail→pass | 20,481 | 10,684 | -48% | 1 | 1 | 0% | 877 | 1,214 | +38% | 0 | 0 | — |
case-14 | fail→pass | 12,944 | 10,234 | -21% | 1 | 1 | 0% | 1,337 | 990 | -26% | 0 | 0 | — |
case-15 | pass→pass | 14,262 | 11,104 | -22% | 1 | 1 | 0% | 1,565 | 1,155 | -26% | 0 | 0 | — |
case-16 | fail→pass | 12,083 | 10,914 | -10% | 1 | 1 | 0% | 1,146 | 1,041 | -9% | 0 | 0 | — |
case-17 | fail→pass | 17,144 | 10,453 | -39% | 1 | 1 | 0% | 618 | 1,129 | +83% | 0 | 0 | — |
case-18 | pass→pass | 15,884 | 10,173 | -36% | 1 | 1 | 0% | 1,671 | 1,166 | -30% | 0 | 0 | — |
case-19 | pass→pass | 21,091 | 19,034 | -10% | 1 | 1 | 0% | 3,113 | 2,935 | -6% | 0 | 0 | — |
case-20 | pass→pass | 15,645 | 15,597 | -0% | 1 | 1 | 0% | 1,794 | 1,851 | +3% | 0 | 0 | — |
case-21 | pass→pass | 12,544 | 9,883 | -21% | 1 | 1 | 0% | 1,352 | 1,023 | -24% | 0 | 0 | — |
case-22 | pass→pass | 17,577 | 19,792 | +13% | 1 | 1 | 0% | 2,071 | 2,468 | +19% | 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 16 counted toward the lift figure. The other 6 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 16 comparable cases. 2 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.