Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →SAFe development workflow guidance including branch naming conventions, commit message format, rebase-first workflow, and CI validation. Use when starting work on a Linear ticket, preparing commits, creating branches, writing PR descriptions, or asking about contribution guidelines.
.claude/skills/bybren-llc-safe-workflow/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -23% | 0% |
> TEMPLATE: This skill uses {{TICKET_PREFIX}} as a placeholder. Replace with your project's ticket prefix (e.g., WOR, PROJ, FEAT).
Enforce SAFe-compliant git workflow with standardized branch naming, commit message format, and rebase-first merge strategy.
Format: {{TICKET_PREFIX}}-{number}-{short-description}
text# Good {{TICKET_PREFIX}}-447-create-safe-workflow-skill {{TICKET_PREFIX}}-123-fix-login-redirect # Bad feature/add-dark-mode (missing ticket number) john-new-feature (personal naming)
Format: type(scope): description [{{TICKET_PREFIX}}-XXX]
| Type | When to Use | | ---------- | ----------------------------- | | feat | New feature | | fix | Bug fix | | docs | Documentation only | | refactor | Code restructuring | | test | Adding or updating tests | | chore | Maintenance, dependencies |
textfeat(harness): create safe-workflow skill [{{TICKET_PREFIX}}-447] fix(auth): resolve login redirect [{{TICKET_PREFIX}}-57]
bash# 1. Start from latest main git checkout {{MAIN_BRANCH}} && git pull origin {{MAIN_BRANCH}} # 2. Create feature branch git checkout -b {{TICKET_PREFIX}}-{number}-{description} # 3. Make commits git commit -m "type(scope): description [{{TICKET_PREFIX}}-XXX]" # 4. Before pushing - rebase git fetch origin && git rebase origin/{{MAIN_BRANCH}} # 5. Push with force-with-lease git push --force-with-lease
{{CI_VALIDATE_COMMAND}}| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | fail→pass | 8,456 | 2,645 | -69% | 1 | 1 | 0% | 1,371 | 983 | -28% | 0 | 0 | — |
case-01 | fail→pass | 8,957 | 4,312 | -52% | 1 | 1 | 0% | 1,277 | 1,353 | +6% | 0 | 0 | — |
case-02 | pass→pass | 10,097 | 4,683 | -54% | 1 | 1 | 0% | 1,752 | 1,365 | -22% | 0 | 0 | — |
case-03 | fail→pass | 15,398 | 6,334 | -59% | 1 | 1 | 0% | 2,292 | 1,701 | -26% | 0 | 0 | — |
case-04 | pass→fail | 10,937 | 5,876 | -46% | 1 | 1 | 0% | 1,764 | 1,482 | -16% | 0 | 0 | — |
case-05 | fail→fail | 14,826 | 11,088 | -25% | 1 | 1 | 0% | 2,530 | 2,512 | -1% | 0 | 0 | — |
case-06 | pass→pass | 7,908 | 4,196 | -47% | 1 | 1 | 0% | 1,341 | 1,228 | -8% | 0 | 0 | — |
case-07 | pass→pass | 19,281 | 13,064 | -32% | 1 | 1 | 0% | 2,870 | 2,538 | -12% | 0 | 0 | — |
case-08 | fail→pass | 6,706 | 3,615 | -46% | 1 | 1 | 0% | 1,122 | 1,146 | +2% | 0 | 0 | — |
case-09 | fail→pass | 8,731 | 2,949 | -66% | 1 | 1 | 0% | 1,394 | 1,072 | -23% | 0 | 0 | — |
case-10 | pass→pass | 9,228 | 3,565 | -61% | 1 | 1 | 0% | 1,434 | 1,101 | -23% | 0 | 0 | — |
case-12 | pass→pass | 5,794 | 2,438 | -58% | 1 | 1 | 0% | 923 | 964 | +4% | 0 | 0 | — |
case-13 | pass→pass | 4,931 | 3,403 | -31% | 1 | 1 | 0% | 764 | 1,041 | +36% | 0 | 0 | — |
case-14 | pass→pass | 6,955 | 3,330 | -52% | 1 | 1 | 0% | 1,144 | 1,069 | -7% | 0 | 0 | — |
case-15 | fail→pass | 6,837 | 3,006 | -56% | 1 | 1 | 0% | 1,067 | 1,027 | -4% | 0 | 0 | — |
case-16 | pass→pass | 6,602 | 3,249 | -51% | 1 | 1 | 0% | 1,006 | 1,060 | +5% | 0 | 0 | — |
case-17 | fail→pass | 12,940 | 2,834 | -78% | 1 | 1 | 0% | 1,959 | 944 | -52% | 0 | 0 | — |
case-18 | pass→pass | 4,141 | 2,537 | -39% | 1 | 1 | 0% | 633 | 957 | +51% | 0 | 0 | — |
case-19 | pass→pass | 6,040 | 3,057 | -49% | 1 | 1 | 0% | 893 | 1,058 | +18% | 0 | 0 | — |
case-20 | pass→pass | 5,996 | 3,332 | -44% | 1 | 1 | 0% | 1,011 | 1,059 | +5% | 0 | 0 | — |
case-21 | fail→pass | 8,093 | 3,425 | -58% | 1 | 1 | 0% | 1,211 | 1,150 | -5% | 0 | 0 | — |
case-22 | fail→pass | 6,049 | 2,794 | -54% | 1 | 1 | 0% | 855 | 1,037 | +21% | 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 +36 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.