📚 Two books on harness engineering — the design philosophies behind Claude Code & Codex: constraints, query loops, context governance, multi-agent verification. harness-books.agentway.dev
Harness Books is a two-volume written resource, available to read online or download as PDFs, that examines the engineering principles behind AI coding agents, with Claude Code as the primary subject of Book 1 and a side-by-side comparison of Claude Code and OpenAI Codex in Book 2. Rather than walking through source code, the books analyze how harness components, specifically the control plane, query loop, tool permissions, context governance via CLAUDE.md and compact operations, failure recovery, and multi-agent verification, work together to keep an agent system bounded and accountable inside real terminal and repository environments. Book 2 focuses on where each system places architectural authority, contrasting Claude Code's runtime discipline against Codex's more structured control layer. The books target developers and engineering teams building or evaluating coding agents, with a particular emphasis on converting individual operational experience into reusable institutional rules. One concrete focus is treating model errors as runtime norms rather than exceptions and designing recovery paths accordingly.
- ✓Recently active
- ✓Healthy fork ratio
- ✓Clear description
- ✓Topics declared
- ✓Documented (README)
- !No standard license detected
git clone https://github.com/wquguru/harness-books && cp harness-books/*.md ~/.claude/agents/Subagents overview
What people ask about harness-books
What is wquguru/harness-books?
+
wquguru/harness-books is subagents for the Claude AI ecosystem. 📚 Two books on harness engineering — the design philosophies behind Claude Code & Codex: constraints, query loops, context governance, multi-agent verification. harness-books.agentway.dev It has 2.5k GitHub stars and was last updated 1mo ago.
How do I install harness-books?
+
You can install harness-books by cloning the repository (https://github.com/wquguru/harness-books) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is wquguru/harness-books safe to use?
+
Our security agent has analyzed wquguru/harness-books and assigned a Trust Score of 75/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.
Who maintains wquguru/harness-books?
+
wquguru/harness-books is maintained by wquguru. The last recorded GitHub activity is from 1mo ago, with 3 open issues.
Are there alternatives to harness-books?
+
Yes. On ClaudeWave you can browse similar subagents at /categories/agents, sorted by popularity or recent activity.
Deploy harness-books to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/wquguru-harness-books)<a href="https://claudewave.com/repo/wquguru-harness-books"><img src="https://claudewave.com/api/badge/wquguru-harness-books" alt="Featured on ClaudeWave: wquguru/harness-books" width="320" height="64" /></a>More Subagents
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
The agent that grows with you
Java 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Production-ready platform for agentic workflow development.
The agent engineering platform.
🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.