An AI-powered translation assistant that translates markdown, legal, and technical documents while strictly preserving structural formatting and terminology consistency.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install document-translation-assistant
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install document-translation-assistant using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Document Translation Assistant?
The Document Translation Assistant is an advanced automation designed for developers, localizers, and legal teams who require precise document translation without sacrificing structural integrity. As one of the premier Openclaw Skills, this tool handles complex formats including Markdown, HTML, DOCX, and text-layered PDFs, distinguishing between translatable prose and non-translatable elements like code fences, inline tags, tables, and system variables.
By leveraging an automated terminology extraction and glossary management engine, this skill ensures domain-specific accuracy across multiple files. It translates segment-by-segment and reviews the entire document post-translation to resolve context breaks, making it perfect for syncing cross-border developer documentation or legal contracts dynamically.
Document Translation Assistant Use Cases
- Open Source Project Localization: Automatically translate developer READMEs, CONTRIBUTING files, and documentation directories from one language to another while preserving URLs, code blocks, and markdown syntax.
- Legal and Compliance Alignment: Translate complex multi-page contracts, NDAs, and terms of service with specialized legal glossary enforcement and side-by-side bilingual formatting.
- Continuous Documentation Syncing: Detect weekly document updates in a repository and incrementally translate only the changed segments, maintaining 100% terminology consistency across versions.
- Multi-Language Package Generation: Deploy localized variants of product documentation (e.g., English, Chinese, Japanese) from a single source document simultaneously.
How Document Translation Assistant Works
- Document Ingestion: The assistant parses the source file (MD, DOCX, PDF, HTML) and maps its elements into a strict structure tree.
- Structure Preservation: It classifies elements as translatable (e.g., body text, headings) or non-translatable (e.g., code blocks, links, inline variables) to protect formatting.
- Terminology Extraction: The engine scans the text to identify domain-specific nouns and phrases depending on the selected mode (Tech, Legal, Marketing, General).
- Glossary Alignment: A candidate list of terms is built, prompting the user to edit, approve, or import translations before the main execution.
- Segmented Translation: Translation occurs in context-optimized chunks with strict glossary enforcement and formatting preservation.
- Consistency and Coherence Auditing: A post-translation scan flags missing terms, mismatched translations, or context breaks across paragraphs.
- Document Generation: The finalized document is compiled into the requested layout, supporting translated-only or dual bilingual outputs.
Document Translation Assistant Setup
To start using this integration among your Openclaw Skills, initialize the assistant and process your first document by running the following commands in your CLI:
# Step 1: Parse the source document to detect structure and language
translation-assistant.sh parse README_zh.md
# Step 2: Extract and verify the terminology glossary
translation-assistant.sh glossary README_zh.md
# Step 3: Translate the document preserving formatting
translation-assistant.sh translate README_zh.md en --domain tech
Document Translation Assistant Data Schema & Taxonomy
The Document Translation Assistant organizes its data structure through persistent local translation glossaries and abstract layout trees:
Glossary JSON Schema
Saves terminology configurations in the local project root (e.g., .translation-glossary.json) for seamless future syncs:
| Key | Type | Description |
|---|---|---|
project |
String | Unique identifier for the translation project |
domain |
String | Chosen domain mode (tech, legal, marketing, general) |
source_lang |
String | ISO language code of the original text |
target_lang |
String | ISO language code of the output text |
terms |
Object | Map of verified source terms to target translations |
Node Classification Schema
Maintains structural integrity using an internal taxonomy:
- Translatable Nodes: Headings, paragraph text, list items, table cell text, and alt texts.
- Preserved Nodes: Code blocks, inline commands, URLs, HTML tags, and bracketed template variables (e.g.,
{{variable}}).
Document Translation Assistant Advanced Features
- Continuous Incremental Syncing: Integrates into Git workflows to diff source documents, translating only modified lines while reusing cached glossaries.
- Format Validation Guardrails: Automatic post-translation structure checks that compare the source vs translated document's block count, halting output on formatting errors.
- Local-Only Offline Privacy: Supports running translations entirely on local models for highly sensitive or proprietary legal documents.
- Side-by-Side Bilingual Compilation: Supports generating dual-column or interleaved bilingual layouts ideal for legal reviews and team collaboration.
- Multi-Language Bulk Publishing: Translates single master assets into multiple target languages simultaneously, complete with locale-specific custom glossaries.
SKILL.md
Loading
Related Openclaw Skills
Goofish Search List Scraper
An automated web scraper skill to extract second-hand product listings, pricing, and seller location data from China's largest used goods marketplace, Goofish (Xianyu).
Goofish Item Detail Extractor
A powerful browser automation skill that extracts complete product listings, pricing, images, and seller details from Goofish (Xianyu) pages.
d-token-saver
A strategic optimization protocol designed to reduce AI context and output token consumption by up to 95% across coding, UI/UX, and creative design pipelines.
Taobao Product Detail
An automated browser-automation skill that navigates Taobao or Tmall pages by itemId to extract comprehensive product data like titles, prices, SKU variants, and images.
Foreign Trade Copywriter
An intelligent multi-language copywriting assistant designed to draft high-converting B2B sales emails, inquiry responses, and product descriptions for global trade.
Mimir Memory
A local, encrypted persistent memory skill for agents via a single Rust binary and MCP server integration.
Featured*
