doc-i18n

Translate FastGPT documentation from Chinese MDX files into North American English.

29.5k|7.3k|Updated Feb 23, 2023
One-click install
npx skills add https://github.com/labring/FastGPT --skill doc-i18n
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: doc-i18n
Source: https://github.com/labring/FastGPT/tree/main/.agents/skills/doc/i18n
Command: npx skills add https://github.com/labring/FastGPT --skill doc-i18n

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Keeping FastGPT's bilingual documentation in sync is tedious: every Chinese .mdx content file and meta.json navigation file needs a matching English counterpart, and inconsistent product terminology across translations confuses users. This Skill automates the translation workflow while enforcing a shared glossary as the single source of truth for product terms.

Core Features & Use Cases

  • Automatic Change Detection: Uses git diff to find changed Chinese .mdx and meta.json files under document/ that need English versions.
  • Glossary-Enforced Translation: Applies longest-match terminology from a shared FastGPT glossary (e.g., 知识库 → Dataset, 数据集 → Collection) and preserves MDX imports, image paths, URLs, JSX components, and code blocks.
  • Navigation Synchronization: Keeps meta.en.json pages arrays aligned with meta.json, translating only titles, descriptions, and separator strings.
  • Use Case: After adding a new Chinese docs page, run this Skill to generate the .en.mdx file, update meta.en.json, and verify no untranslated Chinese text or forbidden terminology remains.

Quick Start

Translate the Chinese documentation files I just modified in the document directory into English and sync the navigation files.

Frequently Asked Questions about doc-i18n

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I translate FastGPT documentation from Chinese to English?

Run this Skill after modifying Chinese docs. It detects changed .mdx and meta.json files via git diff, generates corresponding .en.mdx and meta.en.json files, and applies the shared FastGPT glossary for consistent product terminology.

How are FastGPT product terms translated consistently?

A shared glossary file is the single source of truth, applied with longest-match rules. For example, 知识库 becomes Dataset, 数据集 becomes Collection, and 知识库引用 becomes Dataset citation(s). Forbidden translations listed in the glossary must not be used.

Does the translation preserve MDX components and code blocks?

Yes. MDX import statements, image paths, link URLs, JSX component structure and attributes, table markdown structure, code block contents, and emoji are kept unchanged. Only frontmatter title/description, body text, component text, and Chinese code comments are translated.

What happens to meta.json navigation files during translation?

Each Chinese meta.json gets a synchronized meta.en.json. Only title, description, and separator strings are translated; the pages array, icon, root, and order fields stay identical. New pages missing from navigation are flagged or added after user confirmation.

What are the limitations of automated documentation translation here?

Terms not covered by the shared glossary require user confirmation before being standardized. Navigation placement for new pages that cannot be reliably inferred also requires asking the user rather than guessing.