Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Stage implementation changes for commit with precise file selection. Use when the user asks to "stage changes", "stage files", "add files to staging", or "prepare changes for commit".
.claude/skills/tobihagemann-stage/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -52% | 0% |
Stage implementation changes with precise file selection.
Run git status to see all modified, added, and deleted files.
Stage only the files relevant to the current task:
bashgit add <file1> <file2> ...
git add -A, git add ., or git add -u — all three sweep in unrelated changesgit add -p <file> to stage only the relevant hunks.env, credentials, API keys). Warn if detected.Run git status and git diff --cached to verify the staging area contains exactly the intended changes.
Then call update_plan to mark this step completed and continue with the next step of the active workflow.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 3,158 | 14,297 | +353% | 1 | 1 | 0% | 280 | 357 | +28% | 0 | 0 | — |
case-02 | fail→fail | 4,657 | 3,816 | -18% | 1 | 1 | 0% | 786 | 438 | -44% | 0 | 0 | — |
case-03 | fail→fail | 4,046 | 4,491 | +11% | 1 | 1 | 0% | 677 | 434 | -36% | 0 | 0 | — |
case-04 | fail→fail | 24,596 | 2,934 | -88% | 1 | 1 | 0% | 751 | 634 | -16% | 0 | 0 | — |
case-05 | fail→fail | 7,193 | 2,432 | -66% | 1 | 1 | 0% | 1,295 | 602 | -54% | 0 | 0 | — |
case-06 | pass→pass | 5,316 | 8,662 | +63% | 1 | 1 | 0% | 1,006 | 1,854 | +84% | 0 | 0 | — |
case-07 | pass→pass | 7,547 | 3,939 | -48% | 1 | 1 | 0% | 1,503 | 959 | -36% | 0 | 0 | — |
case-08 | pass→pass | 7,231 | 4,384 | -39% | 1 | 1 | 0% | 1,414 | 859 | -39% | 0 | 0 | — |
case-09 | pass→fail | 4,018 | 4,437 | +10% | 1 | 1 | 0% | 714 | 471 | -34% | 0 | 0 | — |
case-10 | pass→pass | 4,744 | 24,647 | +420% | 1 | 1 | 0% | 821 | 1,308 | +59% | 0 | 0 | — |
case-11 | pass→fail | 5,763 | 5,602 | -3% | 1 | 1 | 0% | 1,044 | 594 | -43% | 0 | 0 | — |
case-12 | pass→pass | 5,623 | 2,788 | -50% | 1 | 1 | 0% | 1,066 | 655 | -39% | 0 | 0 | — |
case-13 | fail→pass | 6,015 | 2,072 | -66% | 1 | 1 | 0% | 1,131 | 530 | -53% | 0 | 0 | — |
case-14 | fail→pass | 3,206 | 1,741 | -46% | 1 | 1 | 0% | 649 | 560 | -14% | 0 | 0 | — |
case-15 | fail→pass | 4,202 | 3,015 | -28% | 1 | 1 | 0% | 902 | 869 | -4% | 0 | 0 | — |
case-16 | pass→pass | 7,665 | 3,210 | -58% | 1 | 1 | 0% | 1,496 | 860 | -43% | 0 | 0 | — |
case-17 | pass→pass | 5,148 | 3,033 | -41% | 1 | 1 | 0% | 963 | 785 | -18% | 0 | 0 | — |
case-18 | fail→pass | 5,823 | 3,327 | -43% | 1 | 1 | 0% | 1,176 | 854 | -27% | 0 | 0 | — |
case-19 | fail→pass | 7,585 | 2,681 | -65% | 1 | 1 | 0% | 1,532 | 738 | -52% | 0 | 0 | — |
case-20 | pass→fail | 8,238 | 3,544 | -57% | 1 | 1 | 0% | 1,608 | 742 | -54% | 0 | 0 | — |
case-21 | pass→pass | 7,497 | 6,512 | -13% | 1 | 1 | 0% | 1,543 | 1,413 | -8% | 0 | 0 | — |
case-22 | pass→pass | 5,176 | 3,730 | -28% | 1 | 1 | 0% | 1,022 | 910 | -11% | 0 | 0 | — |
case-23 | pass→pass | 7,908 | 4,442 | -44% | 1 | 1 | 0% | 1,199 | 1,109 | -8% | 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 18 counted toward the lift figure. The other 5 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 +9 percentage points is the difference between those two pass rates over the 18 comparable cases. 3 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/21/2026 | +9% |
Other measured skills in the registry, with their headline benchmark lift.