Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Branch, deploy, and manage MySQL databases with deploy requests. Use when an agent needs to manage PlanetScale databases, create branches, or submit deploy requests for schema changes.
.claude/skills/composio-community-planetscale-cli-pscale/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -72% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -65% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -57% | 0% |
Branch, deploy, and manage MySQL databases with deploy requests.
bashbrew install planetscale/tap/pscale pscale auth login
bashpscale database list --json pscale branch list <database> --json pscale branch create <database> feature-branch pscale deploy-request create <database> feature-branch pscale deploy-request deploy <database> <number> pscale shell <database> <branch> pscale connect <database> <branch> --port 3306
--json for structured output| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 7,107 | 2,587 | -64% | 1 | 1 | 0% | 1,055 | 406 | -62% | 0 | 0 | — |
case-02 | pass→pass | 159,641 | 9,564 | -94% | 1 | 1 | 0% | 1,121 | 589 | -47% | 0 | 0 | — |
case-03 | fail→pass | 36,984 | 34,449 | -7% | 1 | 1 | 0% | 1,445 | 410 | -72% | 0 | 0 | — |
case-04 | pass→pass | 6,330 | 2,616 | -59% | 1 | 1 | 0% | 742 | 513 | -31% | 0 | 0 | — |
case-05 | pass→pass | 3,443 | 2,096 | -39% | 1 | 1 | 0% | 608 | 535 | -12% | 0 | 0 | — |
case-06 | pass→pass | 5,638 | 2,215 | -61% | 1 | 1 | 0% | 953 | 498 | -48% | 0 | 0 | — |
case-07 | pass→pass | 4,197 | 1,815 | -57% | 1 | 1 | 0% | 617 | 421 | -32% | 0 | 0 | — |
case-08 | pass→pass | 19,279 | 1,365 | -93% | 1 | 1 | 0% | 530 | 384 | -28% | 0 | 0 | — |
case-09 | fail→pass | 49,180 | 48,141 | -2% | 1 | 1 | 0% | 1,145 | 526 | -54% | 0 | 0 | — |
case-10 | fail→pass | 6,844 | 1,920 | -72% | 1 | 1 | 0% | 1,158 | 405 | -65% | 0 | 0 | — |
case-11 | pass→pass | 4,142 | 2,142 | -48% | 1 | 1 | 0% | 428 | 441 | +3% | 0 | 0 | — |
case-12 | pass→pass | 9,430 | 2,770 | -71% | 1 | 1 | 0% | 1,291 | 617 | -52% | 0 | 0 | — |
case-13 | pass→pass | 4,779 | 1,895 | -60% | 1 | 1 | 0% | 836 | 481 | -42% | 0 | 0 | — |
case-14 | pass→pass | 2,589 | 2,948 | +14% | 1 | 1 | 0% | 367 | 525 | +43% | 0 | 0 | — |
case-15 | pass→pass | 6,932 | 2,244 | -68% | 1 | 1 | 0% | 921 | 452 | -51% | 0 | 0 | — |
case-16 | pass→pass | 2,628 | 2,841 | +8% | 1 | 1 | 0% | 426 | 416 | -2% | 0 | 0 | — |
case-17 | pass→pass | 3,266 | 1,651 | -49% | 1 | 1 | 0% | 373 | 415 | +11% | 0 | 0 | — |
case-18 | pass→pass | 4,187 | 98,872 | +2261% | 1 | 1 | 0% | 653 | 401 | -39% | 0 | 0 | — |
case-19 | fail→pass | 35,842 | 4,368 | -88% | 1 | 1 | 0% | 1,636 | 700 | -57% | 0 | 0 | — |
case-20 | pass→pass | 5,458 | 3,350 | -39% | 1 | 1 | 0% | 967 | 727 | -25% | 0 | 0 | — |
case-21 | pass→pass | 31,602 | 7,384 | -77% | 1 | 1 | 0% | 1,541 | 1,384 | -10% | 0 | 0 | — |
case-22 | pass→pass | 10,997 | 2,496 | -77% | 1 | 1 | 0% | 815 | 561 | -31% | 0 | 0 | — |
case-23 | pass→pass | 7,060 | 8,372 | +19% | 1 | 1 | 0% | 1,188 | 1,141 | -4% | 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. 23 cases were attempted. The headline lift of +22 percentage points is the difference between those two pass rates over the 23 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.