From pda
Performs comprehensive code review across 5 dimensions (security, architecture, logic, performance, readability) using parallel agents. Aggregates findings into a unified report with prioritized verdicts. Use when asked to review code, review a PR, or perform a code review.
npx claudepluginhub spaghetti-lover/pda-plugins --plugin pdaThis skill uses the workspace's default tool permissions.
Orchestrates 5 parallel review agents, then aggregates findings into a unified report.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Automates semantic versioning and release workflow for Claude Code plugins: bumps versions in package.json, marketplace.json, plugin.json; verifies builds; creates git tags, GitHub releases, changelogs.
Orchestrates 5 parallel review agents, then aggregates findings into a unified report.
git diff main...HEADgit diff HEAD~1git diff --cached + git diffCollect file list and diffs to pass to each agent.
Launch all simultaneously via the Agent tool. Each agent reads its methodology file first, then reviews the files.
Agent 1 — Security: Read skills/reviewing-security/SKILL.md, follow its checklist. End with PASS/WARN/FAIL.
Agent 2 — Architecture: Read skills/reviewing-architecture/SKILL.md, follow its checklist. End with top 1-3 improvements.
Agent 3 — Logic: Read skills/reviewing-logic/SKILL.md, follow its checklist. End with CLEAN/NEEDS_FIXES/RISKY.
Agent 4 — Performance: Read skills/reviewing-performance/SKILL.md, follow its checklist. End with FAST/ACCEPTABLE/NEEDS_OPTIMIZATION.
Agent 5 — Readability: Read skills/reviewing-readability/SKILL.md, follow its checklist. End with PASS/WARN/FAIL.
Each agent prompt:
You are a {dimension} review agent.
FIRST: Read skills/reviewing-{dimension}/SKILL.md — follow its checklist and output format exactly.
Files to review: {file_list}
Diff context: {diff_summary}
Deduplicate: merge findings that flag the same code location, use the more severe classification.
Priority order:
Output directly to user:
# Code Review Report
**Scope**: {files} | **Date**: {date}
| Dimension | Status | Findings |
|-------------|---------------------------|----------|
| Security | PASS/WARN/FAIL | X |
| Architecture | assessment | X |
| Logic | CLEAN/NEEDS_FIXES/RISKY | X |
| Performance | FAST/ACCEPTABLE/NEEDS_OPT| X |
| Readability | PASS/WARN/FAIL | X |
**Overall**: APPROVE / REQUEST_CHANGES / NEEDS_DISCUSSION
## Critical (fix before merge)
## Important (should fix)
## Suggestions (nice to have)
## Detailed Findings by Dimension
User can customize: "review only security and performance", "review with opus", "review PR #123".