translate-book skill
A skill that translates whole books in PDF, DOCX and EPUB into any language using parallel subagents
- 2,402 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Runs conversion scripts and reads files on disk
- Consumes tokens across parallel subagents
Install
- Works in
Manual install
npx skills add deusyu/translate-book -a claude-code -gNeeds Calibre (ebook-convert), Python, pandoc and the pypandoc pip package.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Runs conversion scripts and reads files on disk
- Consumes tokens across parallel subagents
This is third-party code. Review the repository files before installing.
What it does
The skill translates whole books: it parses the source file into markdown, splits it into chunks, translates them with parallel subagents and reassembles the result into HTML, DOCX, EPUB or PDF. It builds a glossary of terms and names to keep the translation consistent across the book, and can resume an interrupted run. It works in Codex, Claude Code and OpenClaw, with Calibre handling format conversion and Python helper scripts.
Who it is for. For translators, editors and readers who need a draft translation of a whole book.
Good fit when
- You need to translate a PDF, DOCX or EPUB book into another language
- You need a consistent glossary across the whole text
- You need to resume an interrupted translation of large files
Not a fit when
- You need a certified professional translation without human review
- You have no Calibre and Python for the scripts
Example request
Translate this EPUB book into English and build a single glossary of namesLimitations
It needs Calibre with ebook-convert, Python and pandoc, plus beautifulsoup4 for a better table of contents. The output is a machine translation that deserves review. Token usage on long books is significant because of parallel subagents.
How to disable. Remove the translate-book folder from your agent's skills directory, for example ~/.claude/skills/translate-book or ~/.agents/skills/translate-book.
FAQ
What else do I install?
Calibre with ebook-convert, Python, pandoc and the pypandoc pip package, optionally beautifulsoup4.
Can I resume a run?
Yes, the skill keeps intermediate chunks and metadata and can continue the translation.
Discussion
No questions or reviews yet.
Related
SkillsA skill translates an academic paper PDF into a new PDF whose pages match the original one to one, with formulas in LaTeX.
BookTrans translates a whole book in one run: scouting, context-aware translation, editing, footnotes and building to epub or fb2.
Skill that translates English and French course handouts into Chinese and builds HTML, DOCX or Markdown with terminology.
A skill for Chinese-to-English paper translation: the author approves terms first, then it translates and reviews independently
Turns a technical book or a folder of documents into a structured skill with chapters, a glossary and a cheatsheet
A skill for translating whole files with the layout kept: PDF, DOCX, PPTX, XLSX, EPUB, subtitles, images, audio and video.
README in short
The README describes an agent skill for translating books in Codex, Claude Code and OpenClaw. The pipeline takes the source file through markdown chunks to translated chunks and reassembles HTML, DOCX, EPUB and PDF outputs. Install via npx skills add or by cloning into a skills directory. It needs Calibre, Python and pandoc, and its scripts merge metadata and maintain a term glossary. MIT licensed.
README badge
Original SKILL.md
--- name: translate-book description: Translate books (PDF/DOCX/EPUB) into any language using parallel sub-agents. Converts input -> Markdown chunks -> translated chunks -> HTML/DOCX/EPUB/PDF. allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Agent, AskUserQuestion --- # Book Translation Skill You are a book translation assistant. You translate entire books from one language to another by orchestrating a multi-step pipeline. ## Workflow ### 1. Collect Parameters Determine the following from the user's message: - file_path: Path to the input file (PDF, DOCX, or EPUB) - REQUIRED - target_lang: Target language code (default: zh) - concurrency: Number of parallel sub-agents per batch (default: 8) ### 2. Preprocess - Convert to Markdown Chunks Run the conversion script to produce chunks.