Skip to main content

Contribute a Chapter

Use this guide when proposing a new chapter or making a major improvement to an existing one.

What makes a strong chapter​

A strong chapter is:

  • practical
  • tied to a real project or system boundary
  • useful to both learners and working engineers
  • clear about what is beginner-friendly versus production-grade

Preferred chapter structure​

Most substantial chapters should include:

  1. What the concept is
  2. Why it matters in production
  3. One practical architecture or workflow
  4. One Mermaid diagram
  5. One project tie-in
  6. One exercise or implementation prompt

Best chapter topics​

  • production-grade prompt contracts
  • multi-agent orchestration tradeoffs
  • citations and retrieval quality evaluation
  • MCP permissions and audit boundaries
  • operator-facing AI UI patterns
  • observability for traces, cost, and quality

What to avoid​

  • generic "AI 101" explanations with no implementation angle
  • inflated claims about business impact or adoption
  • framework sprawl without a clear reason
  • pages that describe a pattern but never show where it fits in the repo

Fast contribution process​

  1. Open a New chapter proposal issue.
  2. Explain the learner value and target layer.
  3. Draft the page using Chapter Template.
  4. Add the page to sidebars.ts.
  5. Run npm run build.
  6. Link the chapter to a relevant project whenever possible.

Good chapter ideas​

  • evaluation scorecards for agent systems
  • trace review workflow for operators
  • real MCP transport follow-up to Project 04
  • browser agent safety patterns beyond dry-run mode
  • Angular AI state management patterns