Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Draft release notes from merged PRs, propose a semver bump, and emit a copy-pasteable `gh release create` command. Use when preparing a tagged GitHub release.
.claude/skills/wesammustafa-git-release/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-08 | ✓→✗ | ▼ Worse | -8% | 0% |
| case-15 | ✓→✗ | ▼ Worse | 25% | 0% |
| case-18 | ✓→✗ | ▼ Worse | 132% | 0% |
feat: commits)fix:)refactor:, perf:, docs:, chore:, …)BREAKING CHANGE: footers or !: markers)bashgh release create v<version> --title "<title>" --notes-file release-notes.md
When the user is preparing a tagged release. Triggers include:
Ask the user to confirm the proposed version bump before running anything destructive (tag creation, release publish).
gh CLI must be authenticated.v0.1.0.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 12,391 | 7,395 | -40% | 1 | 1 | 0% | 2,246 | 1,586 | -29% | 0 | 0 | — |
case-02 | fail→fail | 6,235 | 4,121 | -34% | 1 | 1 | 0% | 1,022 | 841 | -18% | 0 | 0 | — |
case-03 | pass→pass | 5,662 | 4,075 | -28% | 1 | 1 | 0% | 998 | 1,059 | +6% | 0 | 0 | — |
case-04 | pass→pass | 3,613 | 3,426 | -5% | 1 | 1 | 0% | 685 | 992 | +45% | 0 | 0 | — |
case-05 | pass→pass | 2,987 | 3,618 | +21% | 1 | 1 | 0% | 500 | 990 | +98% | 0 | 0 | — |
case-06 | pass→pass | 4,314 | 4,484 | +4% | 1 | 1 | 0% | 757 | 1,127 | +49% | 0 | 0 | — |
case-07 | pass→pass | 7,420 | 3,171 | -57% | 1 | 1 | 0% | 1,357 | 966 | -29% | 0 | 0 | — |
case-08 | pass→fail | 7,130 | 3,759 | -47% | 1 | 1 | 0% | 1,038 | 956 | -8% | 0 | 0 | — |
case-09 | pass→pass | 5,030 | 4,050 | -19% | 1 | 1 | 0% | 806 | 988 | +23% | 0 | 0 | — |
case-10 | fail→fail | 2,960 | 5,166 | +75% | 1 | 1 | 0% | 487 | 1,188 | +144% | 0 | 0 | — |
case-11 | fail→pass | 10,649 | 4,738 | -56% | 1 | 1 | 0% | 1,720 | 1,068 | -38% | 0 | 0 | — |
case-12 | pass→pass | 3,451 | 2,051 | -41% | 1 | 1 | 0% | 603 | 731 | +21% | 0 | 0 | — |
case-13 | fail→fail | 4,968 | 5,468 | +10% | 1 | 1 | 0% | 761 | 1,200 | +58% | 0 | 0 | — |
case-14 | fail→pass | 5,930 | 3,802 | -36% | 1 | 1 | 0% | 876 | 928 | +6% | 0 | 0 | — |
case-15 | pass→fail | 5,253 | 4,902 | -7% | 1 | 1 | 0% | 992 | 1,237 | +25% | 0 | 0 | — |
case-16 | pass→pass | 5,656 | 4,071 | -28% | 1 | 1 | 0% | 966 | 1,071 | +11% | 0 | 0 | — |
case-17 | pass→pass | 3,272 | 4,026 | +23% | 1 | 1 | 0% | 617 | 905 | +47% | 0 | 0 | — |
case-18 | pass→fail | 2,395 | 3,116 | +30% | 1 | 1 | 0% | 410 | 951 | +132% | 0 | 0 | — |
case-19 | pass→pass | 3,096 | 1,474 | -52% | 1 | 1 | 0% | 488 | 607 | +24% | 0 | 0 | — |
case-20 | pass→pass | 8,819 | 8,867 | +1% | 1 | 1 | 0% | 1,574 | 1,971 | +25% | 0 | 0 | — |
case-21 | pass→pass | 8,607 | 5,306 | -38% | 1 | 1 | 0% | 1,538 | 1,246 | -19% | 0 | 0 | — |
case-22 | pass→pass | 6,185 | 4,817 | -22% | 1 | 1 | 0% | 1,078 | 1,198 | +11% | 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 -33 percentage points is the difference between those two pass rates over the 22 comparable cases. 4 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.