Add comprehensive AI coding agent documentation for Next.js development

[Copilot]

For Contributors

Improving Documentation

• Run pnpm prettier-fix to fix formatting issues before opening the PR.
• Read the Docs Contribution Guide to ensure your contribution follows the docs guidelines: https://nextjs.org/docs/community/contribution-guide

What?

Comprehensive AI coding agent documentation suite in .github/agents/ providing complete Next.js development knowledge base.

10 documentation files (176 KB, 6,368 lines):

• system-instructions.md (34 KB) - Technical foundation: rendering pipeline, caching (4 layers), routing, API patterns, auth, performance
• rules.md (20 KB) - 67+ strict development rules for TypeScript, React, Next.js, security, testing
• architecture.md (18 KB) - Repository internals: monorepo structure, build system, SWC/Turbopack, server/client architecture
• workflows.md (18 KB) - Development, testing, deployment, debugging, contributing procedures
• skills.md (14 KB) - 34 skill areas from beginner to expert with progression paths
• agent.md (14 KB) - Behavioral guidelines, task-specific approaches, error handling
• QUICK_REFERENCE.md (9.8 KB) - Common patterns cheat sheet
• README.md (9.8 KB) - Documentation guide and learning paths
• ZUSAMMENFASSUNG.md (9.5 KB) - German summary
• INDEX.md (6.8 KB) - Complete navigation and statistics

Why?

Enables AI coding agents to operate as Next.js fullstack senior developers by providing:

• Complete technical knowledge (App Router, Server Components, caching strategies, data fetching)
• Strict development rules and best practices
• Repository architecture understanding
• Production-tested patterns and workflows
• 100+ working code examples

How?

Knowledge Structure:

• Progressive learning from basics to framework internals
• Modular organization for targeted reference
• Production patterns from Next.js repository analysis
• Cross-referenced content for comprehensive coverage

Coverage:

• Server/Client Components architecture and composition patterns
• 4-layer caching system (Request Memoization, Data Cache, Full Route Cache, Router Cache)
• Data fetching strategies (SSG, SSR, ISR, Streaming)
• API development (Route Handlers, Server Actions, middleware)
• Database integration patterns (Prisma, connection pooling, transactions)
• Authentication systems (NextAuth.js, JWT, RBAC)
• Testing strategies (unit, integration, E2E with Jest/Playwright)
• Performance optimization (Core Web Vitals, bundle analysis, image/font optimization)
• Deployment procedures (Vercel, Docker, CI/CD)

Example usage for AI agents:

// Agent loads system-instructions.md and understands:
// - Server Components are default (no client JS)
// - Client Components only when interactivity needed
// - ISR for static + fresh data

// Agent follows rules.md:
// Rule #6: MUST explicitly declare Client Components
// Rule #20: MUST fetch data on server when possible

// Agent creates optimal implementation:
export default async function Page() {
  // Server Component - zero client JS
  const data = await fetch('https://api.example.com/data', {
    next: { revalidate: 3600 } // ISR pattern
  })
  
  return <DataView data={data} />
}

Original prompt

Erstelle aus dem next.js repo und all seinen files einen detallierten coding agent set up mit skills.md, agent.md, rules file und project instructions/system instructions.
Die files sollen so gut aufgebaut und detalliert sein das jeder coding agent dadurch ein fullstack next.js profi/senior developer ist - auch wenn er sonst total schlecht in coding ist. er soll alle informationen vorliegen haben die man für next js fullstack deveolpment benötigt

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.