// the find
alchaincyf/hermes-agent-orange-book
Hermes Agent 从入门到精通 · 橙皮书系列 · Nous Research 开源 AI Agent 框架实战指南
A practical guide to Nous Research's Hermes Agent, published as PDFs in Chinese and English with a README, and pinned to Hermes v0.16.0. It is aimed at developers who already use Claude Code or Cursor and want to understand Hermes's self-improvement loop, memory layers, and security model. The repo contains the built PDFs and README, not the book's source text.
The errata section is unusually candid: it names the specific wrong claim in section 03 about data harvesting, cites the official FAQ and AGENTS.md that contradict it, and says which of the author's motives still stand. Edition and product versions are stated explicitly, so a reader can tell which Hermes release a claim applies to. The 2.0 edition was rewritten from scratch against v0.16.0 instead of patched, and it moved to MIT with both language PDFs linked directly from the repo.
The book source is not in the repo, only the PDFs and README, and the erratum admits the PDFs have not been regenerated, so the file people download still carries the retracted claim. Contributors cannot review or patch the text as a diff. The only code is tools/add_toc_bookmarks.py, a PDF bookmark helper, so nothing lets a reader check the product claims (the 64 tools, 23 platforms, command behaviour) against a running install. The README also leans on promotion, with follower counts, an App Store ranking, and WorkBuddy links that tell a developer little about whether the guide is worth reading.