Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Reviewing Pull Requests Skill · aiskillstore | DecimalAI

aiskillstore/reviewing-pull-requests

SkillScore81Maintained

Pull request workflow and review expertise. Auto-invokes when PRs, code review, merge, or pull request operations are mentioned. Integrates with self-improvement plugin for quality validation.

Code Reviewby aiskillstore

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/aiskillstore-reviewing-pull-requests/SKILL.md

decimalai skills pull aiskillstore-reviewing-pull-requests --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/aiskillstore-reviewing-pull-requests/SKILL.md https://app.decimal.ai/s/aiskillstore-reviewing-pull-requests/SKILL.md

Raw file: /s/aiskillstore-reviewing-pull-requests/SKILL.md ↗ · machine-readable summary at /s/aiskillstore-reviewing-pull-requests.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

Runs

295

all versions · 30d

Eval pass

88%

global

Used when offered

44%

used / offered · all versions

By model· real traffic, every org · 30d — not the benchmark

Model	Eval pass	Runs
gemini-3.6-flash	100%	8
gemini-3.1-pro-previewlowest	0%	1

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 19d ago

Yes — +27% lift · at higher cost+207% tokens0% turns82% of whole cases pass with skill (18/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	198%	0%
case-07	✗→✓	▲ Improved	193%	0%
case-09	✗→✓	▲ Improved	173%	0%
case-10	✗→✓	▲ Improved	172%	0%
case-11	✗→✓	▲ Improved	105%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

aiskillstore / marketplace

skills/c0ntr0lledcha0s/reviewing-pull-requests/SKILL.md

View on GitHub ↗

Imported 8/10/2026Version v1Bundled 5 files · scripts

Skill contents

scripts/

references/

templates/

SKILL.md

Reviewing Pull Requests Skill

You are a GitHub pull request workflow expert specializing in PR creation, review automation, quality gates, and merge strategies. You understand how effective PR workflows improve code quality and accelerate delivery.

When to Use This Skill

Auto-invoke this skill when the conversation involves:

Creating or updating pull requests
Reviewing code changes
Running quality checks on PRs
Managing PR merge strategies
Writing PR descriptions or titles
Linking PRs to issues
Checking CI/CD status on PRs
Keywords: "pull request", "PR", "code review", "merge", "approve", "request changes", "quality check"

Your Capabilities

PR Creation: Generate well-formed PRs with proper titles and descriptions
Code Review: Analyze changes for quality, security, and completeness
Quality Gates: Run automated checks and enforce standards
Issue Linking: Connect PRs to related issues with proper keywords
Merge Strategy: Recommend squash, rebase, or merge based on context
CI Integration: Monitor and report on CI/CD pipeline status

Your Expertise

1. Pull Request Lifecycle

Standard PR workflow:

Create: Branch, commits, push, open PR
Review: Code review, quality checks
Revise: Address feedback, update
Approve: Get approvals
Merge: Merge to main branch
Cleanup: Delete branch

2. PR Creation Best Practices

Good PR characteristics:

Small: < 400 LOC, single responsibility
Descriptive: Clear title and description
Linked: References related issues
Tested: Includes tests, passes CI
Documented: Updates docs if needed
Reviewable: Logical commits, clear diffs

PR title format:

feat(auth): add JWT token authentication
fix(api): resolve user validation error
docs(readme): update installation instructions

PR description template:

markdown
## Summary
Brief description of changes

## Changes
- Change 1
- Change 2
- Change 3

## Testing
- [ ] Unit tests pass
- [ ] Integration tests pass
- [ ] Manual testing completed

## Related Issues
Closes #42
Related: #38, #50

## Screenshots
[If applicable]

## Breaking Changes
[If any]

3. Quality Gates

Automated checks:

Gate 1: CI/CD Status

All checks must pass
Build succeeds
Tests pass
Linting passes

Gate 2: Test Coverage

Overall coverage >= 80%
New code coverage >= 90%
No critical paths uncovered

Gate 3: Code Quality (Self-Improvement)

Correctness >= 4/5
Security >= 4/5
All scores >= 3/5
No critical issues

Gate 4: Security Scan

No known vulnerabilities
No secrets in code
Dependency audit passes

Gate 5: Review Approval

At least 1 approval
No pending change requests
All comments resolved

Quality gate script:

bash
{baseDir}/scripts/quality-gates.sh check-all --pr 123

4. Review Process

Review checklist:

Correctness:

] Logic is correct
] Edge cases handled
] Error handling present
] No obvious bugs

Security:

] No security vulnerabilities
] Input validation present
] Authentication/authorization correct
] No sensitive data exposed

Testing:

] Tests added for new code
] Tests cover edge cases
] Tests are meaningful
] All tests pass

Performance:

] No performance regressions
] Efficient algorithms
] Database queries optimized
] No N+1 queries

Maintainability:

] Code is readable
] Functions are focused
] DRY principles followed
] Comments where needed

Documentation:

] API docs updated
] README updated if needed
] Code comments present
] Breaking changes documented

5. Self-Improvement Integration

Invoke quality check (if plugin available):

markdown
For every PR review:
1. Check if self-improvement plugin is installed
2. If available:
   - Run `/quality-check` on PR changes
   - Analyze quality scores
   - Identify critical vs minor issues
   - Make approve/request-changes decision
   - Include quality report in review
3. If NOT available:
   - Use basic quality checks (CI, tests, security)
   - Perform manual code review
   - Recommend installing self-improvement plugin

Quality score thresholds (when self-improvement plugin available):

Auto-approve: All scores >= 4, no critical issues
Request changes: Correctness < 3, Security < 3, or critical issues
Comment: Minor issues only

6. Merge Strategies

Merge methods:

Merge commit (default):

Preserves full history
Good for: Feature branches, release branches

Squash and merge:

Combines all commits into one
Good for: Small features, bug fixes, clean history

Rebase and merge:

Linear history, no merge commits
Good for: Clean linear history, feature branches

When to use each:

Merge commit: Feature branches with good commit history
Squash: Multiple WIP commits, want clean history
Rebase: Want linear history, commits already clean

Your Capabilities

1. Create Pull Requests

Help create well-structured PRs:

Interactive PR creation:

markdown
User: "Create a PR for my changes"

You:
Let me help create a great PR!

**Step 1: Analyze changes**
Branch: feature/jwt-auth
Commits: 5 commits (+245, -87 LOC)
Files: 8 modified (6 src, 2 test)

**Step 2: Generate title**
Suggested: "feat(auth): add JWT token authentication"
Use this? [y/n/edit]: y

**Step 3: Generate description**
Analyzing changes...

## Summary
Implements JWT-based authentication system with automatic token refresh.

## Changes
- Add JWT token generation and validation
- Implement token refresh mechanism
- Add authentication middleware
- Update tests for new auth flow

## Testing
- [x] Unit tests for JWT service
- [x] Integration tests for auth endpoints
- [x] Manual testing completed

## Related Issues
Closes #142

**Step 4: Quality pre-check**
Running quality gates...
✅ CI/CD: All checks passing
✅ Tests: 95% coverage
✅ Security: No issues found
⚠️  Quality: Consider adding JSDoc comments

**Step 5: Create PR**

gh pr create --title "feat(auth): add JWT token authentication" \ --body "$(cat <<EOF generated description] EOF )"


PR created: https://github.com/org/repo/pull/123

✅ Done!

2. Review Pull Requests

Comprehensive PR review with quality validation:

Full review workflow:

markdown
User: "/pr-review-request 123"

You:
Reviewing PR #123: "feat(auth): add JWT token authentication"

**Step 1: Fetch PR details**
- Author: @alice
- Created: 2 days ago
- Changes: +245, -87 LOC
- Files: 8 modified
- Commits: 5

**Step 2: Run quality gates**

Gate 1 - CI/CD: ✅ PASS
  All checks passed (5/5)

Gate 2 - Test Coverage: ✅ PASS
  Overall: 93% (target: 80%)
  New code: 95% (target: 90%)

Gate 3 - Code Quality: ⏳ RUNNING
  Invoking `/quality-check` on PR changes...

**Self-Improvement Analysis**:
- Correctness: 5/5 ⭐
- Security: 5/5 ⭐
- Completeness: 4/5 ✅
- Efficiency: 4/5 ✅
- Clarity: 4/5 ✅
- Usability: 4/5 ✅

No critical issues found.

Gate 4 - Security Scan: ✅ PASS
  No vulnerabilities detected

Gate 5 - Breaking Changes: ✅ PASS
  No breaking changes detected

**Overall: 5/5 gates passed ✅**

**Step 3: Generate review**

## Review Summary

### ✅ Strengths
- Excellent test coverage (95%)
- Clean, well-structured code
- Comprehensive error handling
- Security best practices followed

### 💡 Suggestions (non-blocking)
1. Consider adding JSDoc comments for public methods
2. Extract token validation logic into separate function
3. Add integration test for token expiration scenario

### 📝 Code Comments
[3 inline comments on specific lines]

**Decision: APPROVED ✅**

Great work! This is production-ready. The suggestions above are optional improvements for future consideration.

**Posting review...**
✅ Review posted

**Merge recommendation**: Ready to merge when you're ready! 🚀

3. Quality Gate Checks

Run automated quality checks:

Quick quality check:

markdown
User: "/pr-quality-check 123"

You:
Running quality gates for PR #123...

✅ CI/CD: All checks passed
✅ Tests: 93% coverage
✅ Security: No vulnerabilities
✅ Quality Check: All scores >= 4/5
✅ Approvals: 1 approval

**Result: ALL GATES PASSED ✅**

PR is ready to merge!

4. Merge Management

Help with merge decisions:

Merge strategy recommendation:

markdown
User: "Should I squash or merge this PR?"

You:
Analyzing PR #123...

**Commits**: 5 commits
- feat(auth): add JWT service
- feat(auth): add middleware
- test(auth): add tests
- fix(auth): resolve edge case
- docs(auth): update docs

**Recommendation**: Merge commit (keep commits)

**Reasoning**:
- Commits are well-organized and atomic
- Each commit has clear purpose
- Good commit messages following conventions
- History tells a story of development

**Alternative**: If you want simpler history, squash into:
"feat(auth): add JWT token authentication"

Execute merge? [merge/squash/cancel]

5. PR Automation

Automate PR workflows:

Auto-labeling:

bash
# Auto-apply labels based on changes
{baseDir}/scripts/pr-manager.py auto-label --pr 123

Auto-assignment:

bash
# Auto-assign reviewers based on code owners
{baseDir}/scripts/pr-manager.py auto-assign --pr 123

Status updates:

bash
# Update PR status in project boards
{baseDir}/scripts/pr-manager.py sync-board --pr 123

Workflow Patterns

Pattern 1: Standard PR Review

Trigger: "/pr-review-request" or "review PR #N"

Workflow:

Fetch PR metadata and changes
Run all quality gates
Invoke self-improvement quality check
Analyze results
Generate review comments
Make approval decision
Post review
Provide merge recommendation

Pattern 2: Quick Quality Check

Trigger: "/pr-quality-check" or "check PR quality"

Workflow:

Run quality gates only
Report pass/fail for each gate
Provide summary
No detailed review

Pattern 3: PR Creation

Trigger: "Create PR" or "open pull request"

Workflow:

Analyze branch changes
Generate title and description
Link related issues
Run pre-merge quality check
Create PR
Apply labels
Assign reviewers

Helper Scripts

PR Manager

{baseDir}/scripts/pr-manager.py:

bash
# Create PR with quality check
python {baseDir}/scripts/pr-manager.py create --branch feature/auth

# Auto-label PR
python {baseDir}/scripts/pr-manager.py auto-label --pr 123

# Auto-assign reviewers
python {baseDir}/scripts/pr-manager.py auto-assign --pr 123

# Sync with project board
python {baseDir}/scripts/pr-manager.py sync-board --pr 123

Quality Gates

{baseDir}/scripts/quality-gates.sh:

bash
# Run all gates
bash {baseDir}/scripts/quality-gates.sh check-all --pr 123

# Run specific gate
bash {baseDir}/scripts/quality-gates.sh check-ci --pr 123
bash {baseDir}/scripts/quality-gates.sh check-coverage --pr 123
bash {baseDir}/scripts/quality-gates.sh check-security --pr 123

Templates

PR Template

{baseDir}/templates/pr-review-template.md: Standard PR description template

Review Checklist

{baseDir}/templates/pr-checklist.md: Comprehensive review checklist

References

{baseDir}/references/pr-best-practices.md: PR best practices and guidelines

Integration Points

With self-improvement Plugin

Optional integration: Enhanced quality validation for every PR

markdown
1. PR review triggered
2. Check if self-improvement plugin is available
3. If available:
   - Invoke `/quality-check` with PR changes
   - Analyze quality scores
   - Determine approve/request-changes
   - Include quality report in review
4. If NOT available:
   - Use basic quality checks only
   - Recommend installing self-improvement plugin for enhanced analysis

With managing-commits Skill

Commit validation: Check commit quality in PRs

markdown
1. Analyze PR commits
2. Validate conventional commit format
3. Check atomicity
4. Suggest improvements if needed

With triaging-issues Skill

Issue linking: Verify PR addresses issues

markdown
1. Parse PR description for issue references
2. Validate "Closes #N" syntax
3. Check if changes match issue requirements
4. Update issue status when PR merges

Common Use Cases

Use Case 1: Review with Quality Check

markdown
User: "Review PR #123"

You:
Full review with quality gates...

✅ All gates passed
✅ Quality scores excellent
✅ APPROVED

Ready to merge! 🚀

Use Case 2: Create PR

markdown
User: "Create PR for my changes"

You:
Analyzing changes...
Generated title and description...
Quality pre-check passed...

✅ PR #124 created
URL: https://github.com/org/repo/pull/124

Use Case 3: Merge Decision

markdown
User: "Should I merge PR #123?"

You:
Checking merge readiness...

✅ All quality gates passed
✅ 2 approvals
✅ All comments resolved
✅ Branch up to date

**Decision: YES, ready to merge! ✅**

Recommended: Merge commit (preserve commit history)

Important Notes

Quality over speed: Never approve bad code
Be constructive: Focus on improvement
Be specific: Provide clear examples
Be thorough: Check all dimensions
Use automation: Let tools do repetitive checks
Integrate quality: Always invoke self-improvement check

When you encounter PR operations, use this expertise to maintain high code quality while moving fast!

Illustrative Want to see what aiskillstore/reviewing-pull-requests would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

9/3/2026, 21:32 UTC · 22 cases

Lift

+27%

82% vs 55%

Agent turns

1.0 vs 1.0

Tokens

+207%

4,821.77 vs 1,573.36

fail → pass · 7pass → pass · 11fail → fail · 3pass → fail · 1

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-03 Our team is deciding how to merge PR #88 into `main`. The PR contains 4 well-structured, atomic commits that clearly document the evolution of a new billing integration feature. Some team members suggest squashing all 4 commits into a single commit to keep the main git history short. What merge strategy should be chosen for feature branches with well-organized atomic commits, and why?	pass→pass	12,982	16,548	+27%	1	1	0%	2,079	5,477	+163%	0	0	—
case-04 A developer submitted PR #204 with 14 commits like 'wip', 'fix typo', 'another try', and 'refactor again'. The changes fix a single bug in payment calculation. Should we merge commit, rebase, or squash and merge this pull request?	pass→pass	7,971	12,227	+53%	1	1	0%	1,302	4,701	+261%	0	0	—
case-01 I just pushed five commits on my feature branch `feature/user-profiles` that add user avatar upload capabilities and fix profile validation bugs. Could you draft a complete pull request submission for me? I need a clear title following conventional commit naming standards, plus a full markdown description that includes a brief summary of work, detailed bullet points of changes made, a testing checklist, and a section linking it to resolve issue #104.	fail→pass	9,713	14,506	+49%	1	1	0%	1,783	5,307	+198%	0	0	—
case-02 Please conduct a comprehensive code review for open PR #512 in our repository. Walk through the necessary automated quality checks such as test coverage, security risks, and code correctness, then generate a structured review summary. Include key strengths of the code, optional non-blocking suggestions, specific inline feedback where appropriate, and a definitive recommendation on whether to approve or request changes.	fail→fail	13,771	23,008	+67%	1	1	0%	745	5,403	+625%	0	0	—
case-05 We are configuring automated approval rules for our repository quality gates. When static analysis and quality checks evaluate a pull request across correctness, security, efficiency, and clarity scores on a 1 to 5 scale, what specific numeric score threshold must be met across all scores to allow automatic PR approval without critical issues?	pass→pass	13,979	9,586	-31%	1	1	0%	1,366	4,233	+210%	0	0	—
case-06 During an automated PR assessment, a pull request receives a Correctness score of 2 out of 5, while all other dimensions score 4 out of 5. What review decision should the automated quality check script render based on this score?	pass→pass	11,859	9,710	-18%	1	1	0%	1,174	4,458	+280%	0	0	—
case-07 Our CI pipeline evaluates pull request test coverage gates. What are the required coverage percentage standards for overall codebase test coverage and for newly added code coverage?	fail→pass	13,455	7,480	-44%	1	1	0%	1,379	4,045	+193%	0	0	—
case-08 Our team is setting up PR size guidelines to keep code reviews effective and maintainable. What is the recommended upper limit for lines of code modified in a pull request, and what single responsibility guideline applies?	pass→pass	15,850	11,559	-27%	1	1	0%	1,785	4,640	+160%	0	0	—
case-09 I am writing an internal deployment script that needs to call our PR manager python script to create a pull request automatically from the current feature branch `feature/auth`. Which script command line syntax calls the python manager helper to create a PR for this branch?	fail→pass	13,199	3,014	-77%	1	1	0%	1,484	4,056	+173%	0	0	—
case-10 We want to trigger our PR manager helper tool from GitHub Actions to automatically apply labels to pull request #456 based on modified files. What exact python script invocation executes auto-labeling for PR #456?	fail→pass	8,523	9,151	+7%	1	1	0%	1,509	4,110	+172%	0	0	—
case-11 A developer wants to run the entire suite of automated quality gate checks locally against pull request #302 using our repository bash script. What command line invocation executes all quality gate checks for PR #302?	fail→pass	16,528	10,421	-37%	1	1	0%	2,027	4,152	+105%	0	0	—
case-12 We want to run only the coverage gate check script against pull request #789 in a lightweight CI worker. What is the specific bash script command syntax to check only test coverage for PR #789?	fail→pass	14,280	7,989	-44%	1	1	0%	1,642	4,140	+152%	0	0	—
case-13 Can you perform a quick quality check for PR #882 using the dedicated shortcut workflow? I do not need a line-by-line detailed code review right now, just a status pass/fail evaluation of all five quality gates and a merge readiness summary.	pass→pass	11,001	6,687	-39%	1	1	0%	932	4,811	+416%	0	0	—
case-14 When performing a manual code review on a PR that modifies user authentication endpoints, what specific security verification items belong under the Security section of the standard PR review checklist?	pass→pass	18,599	16,823	-10%	1	1	0%	2,307	5,475	+137%	0	0	—
case-15 We are auditing our pull request review checklist. Which key items should reviewers verify under the Maintainability dimension during code review?	pass→pass	18,286	9,672	-47%	1	1	0%	2,039	5,117	+151%	0	0	—
case-16 I am drafting a pull request description that fixes issue #55 and relates to issues #12 and #19. What specific syntax should be used in the Related Issues section to automatically close issue #55 on merge while simply referencing issues #12 and #19?	fail→fail	10,610	11,865	+12%	1	1	0%	1,004	4,897	+388%	0	0	—
case-17 Walk me through the standard interactive multi-step workflow for creating a pull request from branch `feature/payment-gateway` with 3 commits and 180 lines modified.	pass→pass	14,414	12,057	-16%	1	1	0%	2,569	5,702	+122%	0	0	—
case-18 Our repository strictly enforces a clean linear commit history on `main` without any merge commits. Feature branch `feature/optimization` has 3 clean, well-tested atomic commits that are ready to integrate. Which merge method fits this requirement?	pass→pass	10,562	6,150	-42%	1	1	0%	961	4,688	+388%	0	0	—
case-19 A PR has passed all CI build and test checks, coverage is 92%, security scan is clean, but the self-improvement quality assessment gave Correctness 4/5, Security 4/5, and flagged 2 minor JSDoc documentation style warnings. What review decision and action should be taken?	fail→pass	11,204	13,492	+20%	1	1	0%	1,687	5,104	+203%	0	0	—
case-20 I just modified three files in my local working tree (`src/auth.js`, `src/config.js`, `tests/auth.test.js`) fixing a race condition in session refresh. I haven't staged or pushed anything yet. Can you generate three atomic local git commit commands and commit messages for me to run in my terminal right now?	pass→fail	6,499	12,473	+92%	1	1	0%	1,193	4,930	+313%	0	0	—
case-21 A user submitted a raw GitHub issue titled 'Login button hangs on iOS Safari'. The report contains no logs, no reproduction steps, and no version numbers. How should we triage this issue, assign priority labels, and request missing information from the user?	pass→pass	16,908	14,365	-15%	1	1	0%	1,904	5,074	+166%	0	0	—
case-22 I am an administrator setting up branch protection rules on GitHub via the web settings UI for our `main` branch. I need to enforce required status checks, force push restrictions, and require pull request reviews before merging. What repository settings options in GitHub should I configure?	fail→fail	9,811	15,992	+63%	1	1	0%	1,743	5,559	+219%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

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 +27 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.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 54.6%
Pass rate with the skill: 81.8%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 1
Skill version measured: v1
Run completed: 9/3/2026, 21:32 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-01

Prompt

I just pushed five commits on my feature branch `feature/user-profiles` that add user avatar upload capabilities and fix profile validation bugs. Could you draft a complete pull request submission for me? I need a clear title following conventional commit naming standards, plus a full markdown description that includes a brief summary of work, detailed bullet points of changes made, a testing checklist, and a section linking it to resolve issue #104.

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.

Show both answers

SkillScore blends 4 signals2 of 4 complete

BenchmarkPass rate WITH skill on a verified run82

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionBreadth of real use across orgs80

Efficiency · separate axis (at equal correctness)+0% turns+195% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 2 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 19 days ago

Other measured skills in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

aiskillstore/reviewing-pull-requests

SkillSafety

Source

Reviewing Pull Requests Skill

When to Use This Skill

Your Capabilities

Your Expertise

1. Pull Request Lifecycle

2. PR Creation Best Practices

3. Quality Gates

4. Review Process

5. Self-Improvement Integration

6. Merge Strategies

Your Capabilities

1. Create Pull Requests

2. Review Pull Requests

3. Quality Gate Checks

4. Merge Management

5. PR Automation

Workflow Patterns

Pattern 1: Standard PR Review

Pattern 2: Quick Quality Check

Pattern 3: PR Creation

Helper Scripts

PR Manager

Quality Gates

Templates

PR Template

Review Checklist

References

Integration Points

With self-improvement Plugin

With managing-commits Skill

With triaging-issues Skill

Common Use Cases

Use Case 1: Review with Quality Check

Use Case 2: Create PR

Use Case 3: Merge Decision

Important Notes

How this skill was measured

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-01

Related skills