atn-claudecode-config

Audit codebase for AI agent readiness — optimize for automated contributors

Run all audits (code, tests, docs, comments) in parallel waves

Audit codebase for performance, security, and maintainability issues

Audit code comments for relevance, remove unnecessary ones, add missing ones

Audit and manage project dependencies

Audit codebase and update documentation for accuracy

Audit test coverage and create missing tests

Cancel active Ralph Loop

Generate changelog from git history. Creates release notes from commits since last tag.

Dead Code Cleanup

Peer review code changes using OpenAI Codex agents for a second AI opinion

Validate and commit current changes with conventional commit message

Multi-agent debugging engine with hypothesis-driven root cause analysis

Guided feature development with codebase understanding and architecture focus

Fix automated PR review comments (CodeRabbit, CodeScene, etc.)

Initialize Claude configuration for a project. Creates CLAUDE.md and recommended settings.

Performance Analysis

Pull Request Workflow

Prune old and merged feature branches. Safely removes local and remote branches after merge.

Start Ralph Loop in current session

CodeRabbit-style multi-agent code review with inline findings and verdicts

Review an implementation plan for gaps, edge cases, and ambiguities

Safe Rollback

Prime Claude with project context. Run at session start for faster, more accurate responses.

Triage Sentry errors across projects. Investigate, fix, ignore, or resolve issues.

Build, commit, and ship changes to production

Audits code for accessibility (WCAG 2.1). USE WHEN reviewing UI components for a11y, fixing screen reader issues, checking color contrast, keyboard navigation, ARIA attributes, focus management, or ensuring WCAG compliance.

Database optimization expert. USE WHEN optimizing slow queries, designing schemas, planning migrations, fixing N+1 queries, adding indexes, reviewing Prisma/Drizzle/TypeORM models, or debugging database performance.

Performance analysis agent. USE WHEN profiling, finding bottlenecks, optimizing speed, analyzing Core Web Vitals, bundle size, render performance, slow queries, or memory leaks.

Expert at safe code refactoring. USE WHEN restructuring code, renaming symbols, extracting functions/modules, simplifying conditionals, removing dead code, converting callbacks to async/await, or modernizing code without changing behavior.

Security vulnerability scanner. USE WHEN auditing for OWASP vulnerabilities, finding hardcoded secrets, checking dependencies for CVEs, reviewing auth/access control, or scanning for injection/XSS risks.

Create distinctive frontend interfaces, review UI against web guidelines, and build brand identities. USE WHEN building web UI, designing components, pages, landing pages, dashboards, auditing UX/accessibility, reviewing design, creating brand assets, logos, or styling web applications.

Plan, orchestrate, and execute multi-step implementation tasks. USE WHEN writing plans, breaking plans into tasks, executing plans, dispatching parallel agents, subagent-driven development, or finishing development branches.

Reference patterns for common code structures. USE WHEN building API clients, designing REST/GraphQL APIs, implementing error handling, state management, UI components, or database schemas.

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

10 hooks across 3 events