# wquguru/harness-books

📚 Two books on harness engineering — the design philosophies behind Claude Code & Codex: constraints, query loops, context governance, multi-agent verification. harness-books.agentway.dev

Repository: https://github.com/wquguru/harness-books
Canonical: https://ross.abutalabs.com/products/harness-books
Homepage: http://harness-books.agentway.dev/
Language: Python
License Family: other
Topics: agentic-ai, ai-agents, ai-engineering, claude-code, codex, context-engineering, developer-tools, llm, prompt-engineering
Last push: 2026-04-19T03:01:21+00:00

## Health v2 (maintenance only)
Score: 50/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 78, release rhythm 35, longevity 11
- inputs: {"age_days": 154, "days_push": 136, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2902, forks 348 (observed 2026-08-28T04:07:28.931289+00:00)

## What it is
Two open-access books on harness engineering — the design philosophies behind Claude Code and Codex — covering constraints, query loops, permission decisions, context governance, recovery, and multi-agent verification. The repository hosts the book sources (Python tooling) with free online reading and PDF downloads.

## Use cases
- learn how Claude Code's agent harness is designed
- compare Claude Code and Codex architecture philosophies
- understand context governance and prompt layering for AI coding agents
- design permission and approval policies for coding agents
- study agent failure recovery and multi-agent verification patterns
- turn individual AI agent experience into reusable team rules

## When to choose
- you build or evaluate AI coding agent systems and want architectural depth beyond feature checklists
- you need a structured framework for agent control planes, state, and governance
- you want free, readable books on agentic AI engineering in English or Chinese

## When to avoid
- you need hands-on code or a runnable agent framework rather than design philosophy
- you want a line-by-line source code walkthrough of Claude Code or Codex
- you require a licensed, redistributable open-source work (no license is provided)

## Facets
- artifact type: learning-resource
- maturity: active
- function: documentation, prompt-engineering, agent-framework, developer-tools
- domain: artificial-intelligence, large-language-models, developer-tools, tutorials, awesome-lists
- platform: cross-platform
- tags: harness-engineering, claude-code, codex, context-engineering, agentic-ai, books, ai-coding-agents, agent-architecture, ai-agents, web-server

## Member repositories
- wquguru/harness-books (main) score 50

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:28.931289+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:35:08.463113+00:00, confidence not recorded.
  - readme: https://github.com/wquguru/harness-books (fetched 2026-08-28T04:07:28.931289+00:00, sha 78c0671926ab)
  - homepage: http://harness-books.agentway.dev/ (fetched 2026-08-29T09:49:52.888219+00:00, sha 749e14be5ae1)
  - site_page: https://agentway.dev/zh/pricing (fetched 2026-08-29T09:49:52.891035+00:00, sha d99b21dd4f80)
- Data as of 2026-08-30T08:39:29.467469+00:00.