#329329th of 389 in Skills, plugins and prompts
agent-rules-books
AGENTS.md rules and agent skills distilled from classic software books
- Stars
- 3.3k
- License
- MIT
- Last commit
- Sep 2026
- Last release
- May 2026
8 on 21 Apr 20263,251 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Markdown rule sets for coding agents, each based on a book such as Clean Code, Refactoring, Domain-Driven Design or Release It. Every book ships in nano, mini and full sizes, and each directory has a SKILL.md that loads the mini version. Install all skills with `npx skills add ciembor/agent-rules-books --all`, or copy the files into AGENTS.md, CLAUDE.md or Cursor rules.
Who it is for: Developers who want coding agents to follow established design principles
Strengths
- Three sizes per book (nano, mini, full) fit different context budgets
- Per-book install through the Skills CLI using --skill
- Plain Markdown, usable with Codex, Cursor, Claude Code and Copilot
- Docs cover per-editor usage and which books combine well
Weaknesses
- Only evidence is one ChatGPT-rated refactor, which the README calls non-benchmark
- Reek smell count barely moved in that test: 1083 before, 1077 after
- No equivalent experiments for other books are planned
- Unofficial paraphrases, not from the authors or publishers
What it needs
- no GPU
- Needs Node.js with npm (for npx skills CLI)
- Models: Codex, Cursor, Claude Code, GitHub Copilot
Quick answers
Is agent-rules-books open source?
Yes. Its code is on GitHub under the MIT license.
Also in Skills, plugins and prompts
See all 389| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |