Agent Automation & Development Skills

Use when creating a new Agent2 agent, scaffolding a service, or when user says "new agent" or "add agent" — generates schema, tools, config, Dockerfile, and Docker Compose entry following Agent2…

Use when orchestrating multi-agent adversarial code reviews using Claude Code agent teams. Activates when conducting thorough code reviews, setting up review debate teams, spawning specialized review personas, or synthesizing findings from multiple reviewers. Provides spawn prompts, debate protocol, and synthesis templates.

Use when orchestrating multi-agent adversarial planning debates before code is written. Activates when planning complex implementations, coordinating pre-implementation discovery, stress-testing implementation approaches, or producing validated plans for coding agents. Provides discovery prompts, debate protocol, and plan templates.

Use when writing, reviewing, or optimizing React and Next.js code. Provides 47 performance best practices covering async patterns, bundle optimization, server/client rendering, re-render prevention, and JavaScript performance. Triggers on tasks involving React components, hooks, data fetching, code splitting, memoization, or performance improvements.

Use when launching Task agents, batch processing multiple files, delegating work to sub-agents, or running parallel operations. Provides XML prompt structure, agent type selection, and coordination patterns.

Use when running Vitest tests to ensure safe execution. Applies when running tests, executing test suites, or invoking pnpm test, npm test, or vitest commands. Prevents hung processes through timeouts, manages watch mode, and ensures proper cleanup. Essential for automated test runs to avoid runaway processes consuming CPU for hours or days.