Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification
.claude/skills/mkurman-using-git-worktrees/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -26% | 0% |
-----|--------| | .worktrees/ exists | Use it (verify ignored) | | worktrees/ exists | Use it (verify ignored) | | Both exist | Use .worktrees/ | | Neither exists | Check CLAUDE.md → Ask user | | Directory not ignored | Add to .gitignore + commit | | Tests fail during baseline | Report failures + ask | | No package.json/Cargo.toml | Skip dependency install |
git check-ignore before creating project-local worktreeYou: I'm using the using-git-worktrees skill to set up an isolated workspace.
[Check .worktrees/ - exists]
[Verify ignored - git check-ignore confirms .worktrees/ is ignored]
[Create worktree: git worktree add .worktrees/auth -b feature/auth]
[Run npm install]
[Run npm test - 47 passing]
Worktree ready at /Users/jesse/myproject/.worktrees/auth
Tests passing (47 tests, 0 failures)
Ready to implement auth featureNever:
Always:
Called by:
Pairs with:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-12 | fail→pass | 8,335 | 3,026 | -64% | 1 | 1 | 0% | 1,066 | 1,029 | -3% | 0 | 0 | — |
case-01 | fail→fail | 5,120 | 4,804 | -6% | 1 | 1 | 0% | 266 | 824 | +210% | 0 | 0 | — |
case-02 | fail→fail | 5,392 | 6,862 | +27% | 1 | 1 | 0% | 262 | 897 | +242% | 0 | 0 | — |
case-03 | fail→fail | 4,659 | 6,238 | +34% | 1 | 1 | 0% | 181 | 952 | +426% | 0 | 0 | — |
case-04 | fail→pass | 9,749 | 2,298 | -76% | 1 | 1 | 0% | 1,597 | 908 | -43% | 0 | 0 | — |
case-05 | fail→pass | 3,903 | 4,191 | +7% | 1 | 1 | 0% | 624 | 1,072 | +72% | 0 | 0 | — |
case-06 | fail→pass | 6,718 | 3,928 | -42% | 1 | 1 | 0% | 1,065 | 1,251 | +17% | 0 | 0 | — |
case-07 | pass→pass | 9,279 | 3,508 | -62% | 1 | 1 | 0% | 1,609 | 1,175 | -27% | 0 | 0 | — |
case-08 | fail→pass | 11,261 | 4,890 | -57% | 1 | 1 | 0% | 1,802 | 1,329 | -26% | 0 | 0 | — |
case-09 | pass→pass | 7,628 | 2,214 | -71% | 1 | 1 | 0% | 1,266 | 902 | -29% | 0 | 0 | — |
case-10 | pass→pass | 3,794 | 2,171 | -43% | 1 | 1 | 0% | 676 | 896 | +33% | 0 | 0 | — |
case-11 | pass→pass | 7,452 | 2,949 | -60% | 1 | 1 | 0% | 1,308 | 1,042 | -20% | 0 | 0 | — |
case-13 | pass→pass | 5,779 | 2,101 | -64% | 1 | 1 | 0% | 964 | 851 | -12% | 0 | 0 | — |
case-14 | pass→pass | 5,037 | 2,940 | -42% | 1 | 1 | 0% | 787 | 975 | +24% | 0 | 0 | — |
case-15 | pass→pass | 4,306 | 2,643 | -39% | 1 | 1 | 0% | 718 | 1,016 | +42% | 0 | 0 | — |
case-16 | pass→pass | 11,527 | 2,637 | -77% | 1 | 1 | 0% | 1,688 | 967 | -43% | 0 | 0 | — |
case-17 | fail→pass | 8,864 | 2,575 | -71% | 1 | 1 | 0% | 1,353 | 913 | -33% | 0 | 0 | — |
case-18 | pass→pass | 8,244 | 2,571 | -69% | 1 | 1 | 0% | 1,292 | 921 | -29% | 0 | 0 | — |
case-19 | fail→pass | 9,141 | 4,401 | -52% | 1 | 1 | 0% | 1,539 | 1,320 | -14% | 0 | 0 | — |
case-20 | pass→pass | 5,936 | 4,283 | -28% | 1 | 1 | 0% | 1,013 | 1,247 | +23% | 0 | 0 | — |
case-21 | pass→pass | 8,384 | 8,783 | +5% | 1 | 1 | 0% | 1,618 | 1,574 | -3% | 0 | 0 | — |
case-22 | pass→pass | 6,817 | 3,505 | -49% | 1 | 1 | 0% | 1,206 | 1,068 | -11% | 0 | 0 | — |
case-23 | fail→fail | 2,815 | 3,809 | +35% | 1 | 1 | 0% | 420 | 1,166 | +178% | 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, and 20 counted toward the lift figure. The other 3 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 +30 percentage points is the difference between those two pass rates over the 20 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.