Harness Engineering: The Complete Guide
Engineering Practices for the Age of AI Coding — From CLAUDE.md to a Complete Engineering System
What Is This Book?
When AI can write code, what becomes the core human skill? The ability to design harnesses.
A harness is the tool shell that allows an AI agent to affect the real world. If the reasoning model is the brain, the harness is the hands and feet. This book is a comprehensive, practical guide to the emerging discipline of Harness Engineering — the art and science of designing constraints, instructions, feedback loops, memory systems, and orchestration layers that make AI agents reliable.
What's Inside
| Part | Chapters | What You'll Learn |
|---|---|---|
| Origins | §01-§03 | What a harness is, 60-year history, three namings (Prompt → Context → Harness) |
| Framework | §04-§05 | Five components of a harness, the counterintuitive art of subtraction |
| Case Studies | §06-§12 | 7 real-world cases: OpenAI Codex, Mitchell Hashimoto, Anthropic, Stripe Minions, LangChain, Kent Beck, Huashu |
| Hands-on | §13-§17 | Build your first harness: instruction layer, constraint layer, capability & memory, orchestration |
| Reflections | §18 | Who will design the next generation of harnesses |
Key Insights
- Agent = Model + Harness — LangChain's coding agent jumped from Top 30 to Top 5 on Terminal Bench 2.0, model completely unchanged
- One Mistake, One Rule — Mitchell Hashimoto's pragmatic approach: every agent error becomes a permanent prevention
- $9 vs $200 — Anthropic proved single-agent ($9) produces broken output; three-agent harness ($200) produces working applications
- 1,300 PRs/Week — Stripe's Minions: the #1 success factor has nothing to do with AI models — it's the engineering infrastructure built for humans over a decade
- Less Is More — Over-engineered harnesses perform worse than no harness at all. 100-line CLAUDE.md > 1,000-line CLAUDE.md
Downloads
| Version | Language | File |
|---|---|---|
| v260411 | English | Harness-Engineering-The-Complete-Guide-v260411.pdf |
| v260411 | Chinese (中文) | Harness-Engineering-v260411-zh.pdf |
💡 Download the PDF for the best reading experience. GitHub's online preview may not render properly.
About the Author
Huashu — AI Native Coder, indie developer, AI content creator.
- Built Kitten Light (App Store Paid #1, 1M+ users) with zero handwritten code
- Author of "Mastering DeepSeek" (tens of thousands of copies sold)
- 300,000+ followers across platforms
- AI Coding Orange Book series: 7 books published
Find Me
- X/Twitter: @AlchainHust
- YouTube: @Alchain
- Bilibili: Huashu
- WeChat Official Account: Huashu (花叔)
- Website: huasheng.ai
The Orange Book Series
This is the third book in the AI Coding Orange Book series:
- Claude Code: From Beginner to Expert — Practical starter guide
- Claude Code: Source Code Analysis — Anthropic's architecture dissected
- Harness Engineering ← You are here
- Agent Skills — Extending AI capabilities
- OpenClaw — Raise your own AI
- Polymarket Guide — Prediction markets
- Gemma 4 — (In progress)
License
This work is shared for educational purposes. Please credit the author when referencing.
"Don't expect AI to become smart on its own — you need to build it a harness."