github-pr-review

Files (2)

• SKILL.md
• gemini_severity_guide.md · references/

SKILL.md (first lines)

# GitHub PR Review

Resolves Pull Request review comments with severity-based prioritization, fix application, and thread replies.

## Quick Start

```bash
# 1. Check project-specific instructions
cat .claude/CLAUDE.md 2>/dev/null | head -50  # Review project conventions

# 2. Get PR and repo info
PR=$(gh pr view --json number -q '.number')
REPO=$(gh repo view --json nameWithOwner -q '.nameWithOwner')

# 3. Fetch and list comments by severity
gh api repos/$REPO/pulls/$PR/comments | python3 -c "
import json, sys
comments = [c for c in json.load(sys.stdin) if not c.get('in_reply_to_id')]
def sev(b): return 'CRITICAL' if 'critical' in b.lower() else 'HIGH' if 'high' in b.lower() else 'MEDIUM' if 'medium' in b.lower() else 'LOW'
for s in ['CRITICAL','HIGH','MEDIUM','LOW']:
    cs = [c for c in comments if sev(c['body'])==s]
    if cs: print(f'{s} ({len(cs)}): ' + ', '.join(f\"#{c['id]}\" for c in cs))
"

# 4. For each comment: read -> analyze -> fix -> verify -> commit -> reply
# 5. Run tests: make test (or project-specific command)
# 6. Push when all fixes verified
```

## Pre-Review Checklist

Before processing comments, verify:

1. **Project conventions**: Read `.claude/CLAUDE.md`, `.kiro/steering/`, or similar
2. **Commit format**: Check `git log --oneline -5` for project style
3. **Test command**: Identify test runner (`make test`, `pytest`, `npm test`)
4. **Branch status**: `git status` to ensure clean working tree

## Core Workflow

### 1. Fetch PR Comments

```bash
PR=$(gh pr view --json number -q '.number')
REPO=$(gh repo view --json nameWithOwner -q '.nameWithOwner')
gh api repos/$REPO/pulls/$PR/comments > /tmp/pr_comments.json
```

### 2. Classify by Severity

Process in order: CRITICAL > HIGH > MEDIUM > LOW

| Severity | Indicators | Action |
|----------|------------|--------|
| CRITICAL | `critical.svg`, "security", "vulnerability" | Must fix |
| HIGH | `high-priority.svg`, "High Severity" | Should fix |
| MEDIUM | `medium-priority.svg`, "Medium Severity" | Recommended |
| LOW | `low-priority.svg`, "style", "nit" | Optional |

### 3. Process Each Comment

For each comment:

**a. Show context**
```
Comment #123456789 (HIGH) - app/auth.py:45
"The validation logic should use constant-time comparison..."
```

**b. Read affected code and propose fix**

**c. Confirm with user before applying**

**d. Apply fix if approved**

**e. Verify fix addresses ALL issues in the comment**

### 4. Commit Changes

Use **git-commit skill** format for review fixes:

```bash
git add <files>
git commit -m "fix(scope): address review comment #ID

Brief explanation of what was wrong and how it's fixed.
Addresses review comment #123456789."
```

**Review fix commit rules** (see git-commit skill for full details):
- First line: `type(scope): subject` (max 50 chars)
- Types: `fix`, `refactor`, `security`, `test`, `style`, `perf`
- Reference the comment ID in body
- Explain what was wrong and how it's fixed

### 5. Reply to Thread

```bash
COMMIT=$(git rev-parse --short HEAD)
gh api repos/$REPO/pulls/$PR/comments \
  --input - <<< '{"body": "Fixed in '"$COMMIT"'. Replaced set lookup with hmac.compare_digest.", "in_reply_to": 123456789}'
```

**Standard Reply Templates**:

| Situation | Template |
|-----------|----------|
| Fixed | `Fixed in [hash]. [brief description of fix]` |
| Won't fix | `Won't fix: [reason - e.g., out of scope, acceptable risk]` |
| By design | `By design: [explanation of why current behavior is intentional]` |
| Deferred | `Deferred to [issue/task number]. Will address in future iteration.` |
| Acknowledged | `Acknowledged. [brief note, e.g., "acceptable for MVP"]` |

No emojis. Keep it minimal and professional.

### 6. Run Tests

```bash
make test  # or project-specific command

From myths-labs/muse, MIT.

More from myths-labs/muse