Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Merge PR, wait for deploy, verify production health. 10-step workflow: pre-flight, CI check, readiness gate, merge (respects merge queue), detect platform, wait for deploy, canary verification, revert option, deploy report. Auto-detects Fly, Render, Vercel, Netlify, Heroku, GitHub Actions.
.claude/skills/paperclipai-land-and-deploy/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 73% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -32% | 0% |
name: land-and-deploy description: > Merge PR, wait for deploy, verify production health. 10-step workflow: pre-flight, CI check, readiness gate, merge (respects merge queue), detect platform, wait for deploy, canary verification, revert option, deploy report. Auto-detects Fly, Render, Vercel, Netlify, Heroku, GitHub Actions. metadata: sources:
repo: garrytan/gstack path: land-and-deploy/SKILL.md commit: f4bbfaa5bdfd2d6ce59541c2145432febde57fed attribution: Garry Tan license: MIT usage: referenced
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 9,345 | 4,305 | -54% | 1 | 1 | 0% | 1,641 | 674 | -59% | 0 | 0 | — |
case-02 | fail→pass | 6,842 | 13,363 | +95% | 1 | 1 | 0% | 1,178 | 1,371 | +16% | 0 | 0 | — |
case-03 | pass→pass | 7,921 | 11,338 | +43% | 1 | 1 | 0% | 1,188 | 2,057 | +73% | 0 | 0 | — |
case-04 | pass→pass | 7,528 | 3,898 | -48% | 1 | 1 | 0% | 1,285 | 869 | -32% | 0 | 0 | — |
case-05 | pass→pass | 13,078 | 7,591 | -42% | 1 | 1 | 0% | 2,141 | 1,526 | -29% | 0 | 0 | — |
case-06 | pass→pass | 8,956 | 4,744 | -47% | 1 | 1 | 0% | 1,560 | 990 | -37% | 0 | 0 | — |
case-07 | fail→fail | 11,590 | 6,757 | -42% | 1 | 1 | 0% | 1,873 | 1,341 | -28% | 0 | 0 | — |
case-08 | pass→pass | 4,113 | 3,347 | -19% | 1 | 1 | 0% | 622 | 742 | +19% | 0 | 0 | — |
case-09 | pass→pass | 9,329 | 7,345 | -21% | 1 | 1 | 0% | 1,576 | 1,600 | +2% | 0 | 0 | — |
case-10 | pass→pass | 6,027 | 5,530 | -8% | 1 | 1 | 0% | 859 | 1,078 | +25% | 0 | 0 | — |
case-11 | pass→pass | 13,767 | 5,891 | -57% | 1 | 1 | 0% | 2,306 | 1,131 | -51% | 0 | 0 | — |
case-12 | pass→pass | 6,573 | 5,595 | -15% | 1 | 1 | 0% | 1,140 | 1,091 | -4% | 0 | 0 | — |
case-13 | pass→pass | 9,090 | 3,860 | -58% | 1 | 1 | 0% | 1,438 | 780 | -46% | 0 | 0 | — |
case-14 | pass→pass | 9,579 | 4,454 | -54% | 1 | 1 | 0% | 1,520 | 921 | -39% | 0 | 0 | — |
case-15 | pass→pass | 9,498 | 5,793 | -39% | 1 | 1 | 0% | 1,428 | 1,040 | -27% | 0 | 0 | — |
case-16 | pass→pass | 9,466 | 5,729 | -39% | 1 | 1 | 0% | 1,574 | 1,142 | -27% | 0 | 0 | — |
case-17 | fail→pass | 6,755 | 7,761 | +15% | 1 | 1 | 0% | 1,100 | 1,406 | +28% | 0 | 0 | — |
case-18 | pass→pass | 11,595 | 4,816 | -58% | 1 | 1 | 0% | 1,018 | 987 | -3% | 0 | 0 | — |
case-19 | fail→pass | 13,109 | 4,592 | -65% | 1 | 1 | 0% | 2,223 | 1,028 | -54% | 0 | 0 | — |
case-20 | pass→pass | 11,055 | 6,428 | -42% | 1 | 1 | 0% | 1,660 | 946 | -43% | 0 | 0 | — |
case-21 | pass→pass | 6,722 | 6,787 | +1% | 1 | 1 | 0% | 1,421 | 1,658 | +17% | 0 | 0 | — |
case-22 | pass→pass | 7,061 | 4,568 | -35% | 1 | 1 | 0% | 1,229 | 961 | -22% | 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 +14 percentage points is the difference between those two pass rates over the 22 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.