review（审查）

Review Skill - AI Code Review

Perform comprehensive code reviews on staged changes, specific files, or pull requests. Routes to expert agents based on file types and automatically creates tasks for critical issues.

Architecture

review [target] [--focus] [--depth]
    │
    ├─→ Step 1: Determine Scope
    │     ├─ No args → git diff --cached (staged)
    │     ├─ --all → git diff HEAD (all uncommitted)
    │     ├─ File path → specific file diff
    │     └─ --pr N → gh pr diff N
    │
    ├─→ Step 2: Analyze Changes (parallel)
    │     ├─ delta for syntax-highlighted diff
    │     ├─ difft for semantic diff (structural)
    │     ├─ Categorize: logic, style, test, docs, config
    │     └─ Identify touched modules/components
    │
    ├─→ Step 3: Load Project Standards
    │     ├─ AGENTS.md, CLAUDE.md conventions
    │     ├─ .eslintrc, .prettierrc, pyproject.toml
    │     ├─ Detect test framework
    │     └─ Check CI config for existing linting
    │
    ├─→ Step 4: Route to Expert Reviewers
    │     ├─ TypeScript → typescript-expert
    │     ├─ React/JSX → react-expert
    │     ├─ Python → python-expert
    │     ├─ Go → go-expert
    │     ├─ Rust → rust-expert
    │     ├─ Vue → vue-expert
    │     ├─ SQL/migrations → postgres-expert
    │     ├─ Claude extensions → claude-architect
    │     ├─ Multi-domain → parallel expert dispatch
    │     └─ All experts preload: security-ops + testing-ops context
    │
    ├─→ Step 5: Generate Review
    │     ├─ Severity: CRITICAL / WARNING / SUGGESTION / PRAISE
    │     ├─ Line-specific comments (file:line refs)
    │     ├─ Suggested fixes as diff blocks
    │     └─ Overall verdict: Ready to commit? Y/N
    │
    └─→ Step 6: Integration
          ├─ Auto-create tasks (TaskCreate) for CRITICAL issues
          ├─ Link to /save for tracking
          └─ Suggest follow-up: /testgen, /explain

Execution Steps

Step 1: Determine Scope

# Default: staged changes
git diff --cached --name-only

# Check if anything is staged
STAGED=$(git diff --cached --name-only | wc -l)
if [ "$STAGED" -eq 0 ]; then
    echo "No staged changes. Use --all for uncommitted or specify a file."
    git status --short
fi

For PR review:

gh pr diff $PR_NUMBER --patch

For specific file:

git diff HEAD -- "$FILE"

For baseline comparison (--base):

git diff $BASE_BRANCH...HEAD

Step 2: Analyze Changes

Run semantic diff analysis (parallel where possible):

With difft (semantic):

command -v difft >/dev/null 2>&1 && git difftool --tool=difftastic --no-prompt HEAD~1 || git diff HEAD~1

With delta (syntax highlighting):

command -v delta >/dev/null 2>&1 && git diff --cached | delta || git diff --cached

Categorize changes:

git diff --cached --name-only | while read file; do
    case "$file" in
        *.test.* | *.spec.*) echo "TEST: $file" ;;
        *.md | docs/*) echo "DOCS: $file" ;;
        *.json | *.yaml | *.toml) echo "CONFIG: $file" ;;
        *) echo "CODE: $file" ;;
    esac
done

Get diff statistics:

git diff --cached --stat

Step 3: Load Project Standards

# Claude Code conventions
cat AGENTS.md 2>/dev/null | head -50
cat CLAUDE.md 2>/dev/null | head -50

# Linting configs
cat .eslintrc* 2>/dev/null | head -30
cat .prettierrc* 2>/dev/null
cat pyproject.toml 2>/dev/null | head -30

# Test framework detection
cat package.json 2>/dev/null | jq '.devDependencies | keys | map(select(test("jest|vitest|mocha|cypress|playwright")))' 2>/dev/null

Check CI for existing linting:

cat .github/workflows/*.yml 2>/dev/null | grep -E "eslint|prettier|pylint|ruff" | head -10

Step 4: Route to Expert Reviewers

Invoke via Task tool:

Task tool with subagent_type: "[detected]-expert"
model: "sonnet"
Prompt includes:
  - Skill preloading (domain knowledge):
    "First, read these files for review context:
     - Read: skills/security-ops/references/owasp-detailed.md
     - Read: skills/testing-ops/SKILL.md"
  - Diff content
  - Project conventions from AGENTS.md
  - Linting config summaries
  - Requested focus area
  - Request for structured review output

Language-specific preloads (append to the preloading section above):

Step 5: Generate Review

The expert produces a structured review:

# Code Review: [scope description]

## Summary

| Metric | Value |
|--------|-------|
| Files reviewed | N |
| Lines changed | +X / -Y |
| Issues found | N (X critical, Y warnings) |

## Verdict

**Ready to commit?** Yes / No

[1-2 sentence summary of overall quality]

---

## Critical Issues

### `src/auth/login.ts:42`

**Issue:** SQL injection vulnerability in user input handling

**Risk:** Attacker can execute arbitrary SQL queries

**Fix:**

• const query = SELECT * FROM users WHERE id = ${userId};

+ const query = SELECT * FROM users WHERE id = $1; + const result = await db.query(query, [userId]);

---

## Warnings

### `src/components/Form.tsx:89`

**Issue:** Missing dependency in useEffect

**Suggestion:** Add `userId` to dependency array

• useEffect(() => { fetchUser(userId) }, []);

+ useEffect(() => { fetchUser(userId) }, [userId]);

---

## Suggestions

[Style improvements, optional enhancements]

---

## Praise

[Good patterns worth noting]

---

## Files Reviewed

| File | Changes | Issues |
| --- | --- | --- |
| `src/auth/login.ts` | +42/-8 | 1 critical |

Step 6: Integration

Auto-create tasks for CRITICAL issues:

TaskCreate: subject: "Fix: SQL injection in login.ts:42" description: "SQL injection vulnerability found in user input handling." activeForm: "Fixing SQL injection in login.ts:42"

Link with dependencies for related issues:

TaskCreate: #1 "Fix SQL injection in login.ts" TaskCreate: #2 "Fix SQL injection in register.ts" TaskUpdate: taskId: "2", addBlockedBy: ["1"]

After fixing issues:

TaskUpdate: taskId: "1" status: "completed"

Severity System

Focus Modes

Depth Modes

Advanced Flags

--base <branch> - Baseline Comparison

Compare changes against a specific branch instead of HEAD:

/review --base main
/review src/ --base develop --thorough

--json - CI/CD Integration

Output review results as JSON:

{
  "summary": {
    "files_reviewed": 3,
    "lines_changed": { "added": 42, "removed": 8 },
    "issues": { "critical": 1, "warning": 2, "suggestion": 1 }
  },
  "verdict": {
    "ready_to_commit": false,
    "reason": "1 critical issue requires attention"
  },
  "issues": [...]
}

CI/CD usage:

- name: Code Review
  run: |
    claude "/review --json" > review.json
    if jq -e '.issues[] | select(.severity == "critical")' review.json; then
      exit 1
    fi

--fix - Auto-Apply Fixes

Automatically apply suggested fixes:

1. Performs standard review
2. For each fixable issue, prompts for confirmation
3. Uses Edit tool to apply approved fixes
4. Creates TaskUpdate for resolved issues

Non-interactive mode:

/review --fix --auto-approve

CLI Tool Integration

Graceful degradation:

command -v delta >/dev/null 2>&1 && git diff --cached | delta || git diff --cached

Reference Files

For framework-specific checks, see:

• framework-checks.md - React, TypeScript, Python, Go, Rust, Vue, SQL patterns

Integration