Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deploy projects, manage domains, env vars, and inspect logs on Vercel. Use when an agent needs to deploy a web app, configure environment variables, or manage Vercel project settings.
.claude/skills/composio-community-vercel-cli/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -45% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -22% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -43% | 0% |
Deploy projects, manage domains and env vars, inspect logs, and sync local project settings.
bashnpm i -g vercel vercel login
bashvercel deploy vercel deploy --prod vercel env ls vercel env add VARIABLE_NAME production vercel domains ls vercel logs <deployment-url> vercel pull # sync project settings locally vercel inspect <deployment-url>
--token or VERCEL_TOKEN for CI--yes flag to skip confirmation prompts| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 3,560 | 2,243 | -37% | 1 | 1 | 0% | 454 | 517 | +14% | 0 | 0 | — |
case-02 | pass→pass | 157,155 | 3,839 | -98% | 1 | 1 | 0% | 951 | 521 | -45% | 0 | 0 | — |
case-03 | pass→pass | 3,831 | 3,453 | -10% | 1 | 1 | 0% | 674 | 529 | -22% | 0 | 0 | — |
case-04 | pass→pass | 5,907 | 2,872 | -51% | 1 | 1 | 0% | 988 | 568 | -43% | 0 | 0 | — |
case-05 | pass→pass | 4,689 | 2,339 | -50% | 1 | 1 | 0% | 779 | 465 | -40% | 0 | 0 | — |
case-06 | pass→pass | 5,424 | 2,684 | -51% | 1 | 1 | 0% | 932 | 505 | -46% | 0 | 0 | — |
case-07 | pass→pass | 4,552 | 2,601 | -43% | 1 | 1 | 0% | 633 | 512 | -19% | 0 | 0 | — |
case-08 | pass→pass | 9,610 | 3,095 | -68% | 1 | 1 | 0% | 637 | 620 | -3% | 0 | 0 | — |
case-09 | pass→pass | 9,175 | 3,276 | -64% | 1 | 1 | 0% | 800 | 526 | -34% | 0 | 0 | — |
case-10 | fail→pass | 3,451 | 1,349 | -61% | 1 | 1 | 0% | 387 | 346 | -11% | 0 | 0 | — |
case-11 | pass→pass | 26,724 | 2,139 | -92% | 1 | 1 | 0% | 881 | 456 | -48% | 0 | 0 | — |
case-12 | pass→pass | 4,258 | 2,511 | -41% | 1 | 1 | 0% | 604 | 417 | -31% | 0 | 0 | — |
case-13 | fail→fail | 7,930 | 2,558 | -68% | 1 | 1 | 0% | 987 | 494 | -50% | 0 | 0 | — |
case-14 | pass→pass | 5,623 | 2,855 | -49% | 1 | 1 | 0% | 778 | 593 | -24% | 0 | 0 | — |
case-15 | pass→pass | 5,780 | 5,164 | -11% | 1 | 1 | 0% | 997 | 918 | -8% | 0 | 0 | — |
case-16 | pass→pass | 6,110 | 3,530 | -42% | 1 | 1 | 0% | 887 | 612 | -31% | 0 | 0 | — |
case-17 | pass→pass | 5,462 | 2,601 | -52% | 1 | 1 | 0% | 897 | 624 | -30% | 0 | 0 | — |
case-18 | pass→pass | 7,071 | 6,172 | -13% | 1 | 1 | 0% | 829 | 603 | -27% | 0 | 0 | — |
case-19 | pass→pass | 9,522 | 3,256 | -66% | 1 | 1 | 0% | 1,640 | 691 | -58% | 0 | 0 | — |
case-20 | pass→pass | 9,689 | 8,716 | -10% | 1 | 1 | 0% | 1,406 | 1,355 | -4% | 0 | 0 | — |
case-21 | pass→pass | 9,340 | 6,848 | -27% | 1 | 1 | 0% | 1,751 | 1,610 | -8% | 0 | 0 | — |
case-22 | pass→pass | 5,813 | 3,623 | -38% | 1 | 1 | 0% | 984 | 800 | -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. The headline lift of +9 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.