Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Prepare pull request descriptions following project template. Activates on: prepare PR, create PR, pull request, ready for PR, draft PR, write PR
.claude/skills/aiskillstore-pr-prepare/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✗→✓ | ▲ Improved | 70% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -17% | 0% |
Automatically prepare pull request descriptions following project standards.
This skill activates on keywords:
When activated, follow these steps:
Run these commands to understand the changes:
bash# Current state git status # Branch comparison (adjust 'main' if different base branch) git log $(git merge-base HEAD main)..HEAD --oneline # Files changed git diff $(git merge-base HEAD main)..HEAD --stat # Actual diff (for smaller changes) git diff $(git merge-base HEAD main)..HEAD
Identify:
Use this template format:
markdown## Summary [1-3 sentences: what changed and why] ## Changes - **[Component]**: [What changed and why] - **[Component]**: [What changed and why] ## Impact - ✅ [Key benefit or outcome] - ✅ [Another benefit] - ✅ No breaking changes ## Testing - [ ] Tests pass (`uv run pytest`) - [ ] Linting passes (`uv run ruff check`) ## Notes [Optional: known issues, follow-up work, dependencies]
Follow conventional commits format:
| Type | When to Use | |------|-------------| | feat: | New feature | | fix: | Bug fix | | docs: | Documentation only | | refactor: | Code restructuring | | test: | Adding tests | | perf: | Performance improvement | | chore: | Maintenance, dependencies |
Present the complete PR description ready to copy-paste into GitHub.
Remind the user:
@coderabbitai summary placeholdergh pr createUser: "Can you prepare the PR for this branch?"
Claude:
Suggested Title: feat: add user authentication with OAuth2
PR Description:
markdown## Summary Add OAuth2 authentication supporting Google and GitHub providers. ## Changes - **auth/oauth.py**: OAuth2 client implementation with token refresh - **api/routes/auth.py**: Login, logout, and callback endpoints - **models/user.py**: User model with OAuth provider fields ## Impact - ✅ Users can sign in with Google or GitHub - ✅ Secure session management with httponly cookies - ✅ No breaking changes to existing API ## Testing - [x] Tests pass (`uv run pytest`) - [x] Linting passes (`uv run ruff check`) ## Notes Follow-up: Add Microsoft provider support
Ready to copy! Push with:
bashgit push -u origin HEAD gh pr create --fill
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 8,468 | 10,987 | +30% | 1 | 1 | 0% | 365 | 955 | +162% | 0 | 0 | — |
case-02 | fail→fail | 4,941 | 5,396 | +9% | 1 | 1 | 0% | 797 | 1,122 | +41% | 0 | 0 | — |
case-03 | fail→fail | 8,016 | 9,948 | +24% | 1 | 1 | 0% | 439 | 994 | +126% | 0 | 0 | — |
case-08 | pass→pass | 10,213 | 7,736 | -24% | 1 | 1 | 0% | 897 | 1,178 | +31% | 0 | 0 | — |
case-04 | pass→pass | 10,118 | 10,563 | +4% | 1 | 1 | 0% | 1,815 | 2,489 | +37% | 0 | 0 | — |
case-05 | pass→pass | 9,638 | 6,307 | -35% | 1 | 1 | 0% | 1,553 | 1,876 | +21% | 0 | 0 | — |
case-06 | pass→pass | 9,900 | 8,413 | -15% | 1 | 1 | 0% | 769 | 1,372 | +78% | 0 | 0 | — |
case-07 | pass→pass | 12,651 | 8,919 | -29% | 1 | 1 | 0% | 1,247 | 1,479 | +19% | 0 | 0 | — |
case-09 | pass→pass | 18,261 | 3,461 | -81% | 1 | 1 | 0% | 910 | 1,285 | +41% | 0 | 0 | — |
case-10 | pass→pass | 9,390 | 7,544 | -20% | 1 | 1 | 0% | 656 | 1,178 | +80% | 0 | 0 | — |
case-11 | pass→pass | 4,914 | 2,680 | -45% | 1 | 1 | 0% | 876 | 1,175 | +34% | 0 | 0 | — |
case-12 | pass→pass | 10,281 | 4,027 | -61% | 1 | 1 | 0% | 897 | 1,362 | +52% | 0 | 0 | — |
case-21 | fail→pass | 5,715 | 4,327 | -24% | 1 | 1 | 0% | 865 | 1,468 | +70% | 0 | 0 | — |
case-13 | fail→pass | 4,315 | 8,253 | +91% | 1 | 1 | 0% | 716 | 1,193 | +67% | 0 | 0 | — |
case-14 | fail→pass | 9,238 | 4,254 | -54% | 1 | 1 | 0% | 1,487 | 1,404 | -6% | 0 | 0 | — |
case-15 | fail→pass | 11,637 | 1,774 | -85% | 1 | 1 | 0% | 1,213 | 1,018 | -16% | 0 | 0 | — |
case-16 | fail→pass | 7,765 | 2,155 | -72% | 1 | 1 | 0% | 1,337 | 1,108 | -17% | 0 | 0 | — |
case-17 | pass→pass | 16,036 | 2,770 | -83% | 1 | 1 | 0% | 1,692 | 1,205 | -29% | 0 | 0 | — |
case-18 | pass→pass | 8,377 | 7,900 | -6% | 1 | 1 | 0% | 534 | 1,196 | +124% | 0 | 0 | — |
case-19 | pass→pass | 4,959 | 7,900 | +59% | 1 | 1 | 0% | 873 | 1,208 | +38% | 0 | 0 | — |
case-20 | pass→pass | 11,929 | 9,691 | -19% | 1 | 1 | 0% | 2,055 | 1,538 | -25% | 0 | 0 | — |
case-22 | fail→fail | 14,312 | 5,385 | -62% | 1 | 1 | 0% | 1,558 | 1,745 | +12% | 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, and 19 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 +23 percentage points is the difference between those two pass rates over the 19 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.