Skip to content
cccskills
FreePublished on GitHub

better-translate

Best practices for AI-driven English-to-Chinese translation. This skill should be used when the user asks to "translate to Chinese", "update the Chinese translation", "improve Chinese translation", "fix translation quality", "review Chinese translation", or when translating any English text into Chinese. Also applies when polishing an existing Chinese translation of English content.

The English summary is being prepared. Showing the original description.

See how to install

Files included (2)

  • SKILL.md5.2 KB
  • references/translation-examples.md4.4 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

English-to-Chinese Translation

Produce natural, accurate Chinese translations that read as if written by a native speaker — free of 翻译腔 (translationese) and AI slop patterns.

Overview

AI models produce predictable defects in Chinese translation: English word order calqued into Chinese, filler phrases added for politeness, over-emphasis not present in the source, and tone shifts. This skill provides a structured iterative workflow to eliminate these issues through comprehension, translation, and repeated polish passes.

When to Use

  • Translating documents, articles, or prose from English to Chinese
  • Improving or polishing an existing Chinese translation
  • Reviewing machine-translated Chinese content for quality

When NOT to Use

  • Translating into languages other than Chinese
  • Translating non-English source text
  • Writing original Chinese content (not a translation)
  • Quick gloss translations where quality is unimportant

Translation Workflow

Phase 1: Comprehension

Read the full English source document before translating anything.

Identify ambiguous passages, domain-specific terminology, and phrases that could be interpreted multiple ways. Present ambiguities in a table:

English PassagePossible MeaningsChosen Interpretation
.........

For unresolved terminology:

  1. Explore the project context (README, codebase, related docs) for usage conventions.
  2. If ambiguity persists, stop and ask the user.

Do not proceed to translation until all ambiguities are resolved.

Phase 2: Initial Translation

Translate the full document in one pass. During the initial draft:

  • Preserve all formatting (headings, lists, code blocks, links, images)
  • Keep proper nouns untranslated (product names, project names, person names)
  • Translate technical terms by established convention (e.g., "machine learning" → "机器学习")
  • Match the source tone and formality level
  • Do not polish during this phase — produce a complete draft first

Phase 3: Iterative Polish

Run a minimum of 3 rounds. Each round performs three distinct checks in order. If any round produces changes, run one additional round after the minimum 3.

Accuracy Check

Compare translated text against the English source line by line. Fix:

  • 含义改变 — meaning differs from source
  • 添油加醋 — content added that source does not contain
  • 遗漏 — content present in source but missing from translation
  • 格式改变 — formatting broken (headings, links, code blocks)
  • 标点符号 — English punctuation left in Chinese text, or wrong Chinese punctuation
  • 语法错误 — grammatical errors in Chinese

Fluency Check

Read the Chinese text without referring to the English source. Fix:

  • 英语语序 — English word order literally transplanted into Chinese
  • 费解的表述 — phrasing that requires re-reading to understand
  • 缺少衔接 — missing transitions between sentences or paragraphs
  • 不自然的搭配 — word combinations no native Chinese speaker would use
  • 能否改用更自然的表述 — rephrase for naturalness without changing meaning

AI Slop Check

Detect and remove AI-specific defect patterns:

  • 过度强调 — excessive emphasis ("极其", "至关重要" where source says "important")
  • 废话填充 — filler phrases not in source ("值得注意的是", "不言而喻", "毋庸置疑")
  • 不自然的客套 — unnatural hedging or politeness not matching source tone
  • 原文没有的展开 — elaboration added beyond what source states
  • 翻译腔 — literal calques, unnatural collocations, mechanical "的" chains

For a detailed catalog of these patterns with examples, consult references/translation-examples.md.

Final Validation

After iterative polish converges (a round produces zero changes):

  1. Read the complete Chinese document from start to finish, as a Chinese reader, without referring to the English source.
  2. Ask: does this read like a native Chinese speaker wrote it? Any AI slop indicators or 翻译腔 remaining?
  3. If any issue is found, return to Phase 3 for another round.
  4. Claim translation complete only when the full read-through raises no issues.

Do not rush this step.

Quick Reference

PhaseActionGoal
1. ComprehensionRead source, list ambiguities, resolve terminologyFull understanding
2. Initial TranslationTranslate in one pass, preserve formattingComplete draft
3. Iterative PolishAccuracy → Fluency → AI Slop, minimum 3 roundsHigh quality
4. Final ValidationFull read-through as Chinese readerConfirm quality

Additional Resources

  • references/translation-examples.md — Good vs bad translation examples, 翻译腔 pattern catalog, and detailed quality checklist

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Browser automation CLI for AI agents. Use when needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. TRIGGER when user requests to "open a website", "fill out a form", "click a button", "take a screenshot", "debug this in browser", "scrape data from a page", "test this web app", "login to a site", "frontend UI/UX aesthetics", "automate browser actions", or any task requiring programmatic web interaction.

The English summary is being prepared. Showing the original description.

archibate/dotfiles-opencode108Updated Apr 29, 2026

Review common AI slops of defensive programming patterns, avoid silent errors. TRIGGER when reviewing code for defensive anti-patterns, writing fail-fast code, or auditing error handling quality.

The English summary is being prepared. Showing the original description.

archibate/dotfiles-opencode108Updated Apr 29, 2026

Guide for writing ast-grep rules to perform structural code search and analysis. This skill should be used when users need to search codebases using Abstract Syntax Tree (AST) patterns, find specific code structures, or perform complex code queries that go beyond simple text search, or when a simple grep/glob search is insufficient for structural code pattern matching.

The English summary is being prepared. Showing the original description.

archibate/dotfiles-opencode108Updated Apr 29, 2026

This skill should be used when the user asks to "use bilibili API", "download bilibili video", "get bilibili user info", "list bilibili favorites", "send bilibili danmaku", "upload video to bilibili", "monitor bilibili live room", "search bilibili", "get bilibili comments", or needs guidance on the bilibili_api Python library usage, authentication, API endpoints, or workflow patterns.

The English summary is being prepared. Showing the original description.

archibate/dotfiles-opencode108Updated Apr 29, 2026

This skill should be used when sending images, files, or notifications back to the user via messaging platforms (Discord, Feishu, Telegram, etc.) through cc-connect. TRIGGER when agent generates a plot/chart/screenshot and wants to show the user; agent creates a report/PDF/file the user should receive; agent needs to proactively notify the user (e.g. task completed, alert, reminder); user asks to "send image", "show me the chart", "notify me", "send the file", "send to Telegram", "show plot in Discord".

The English summary is being prepared. Showing the original description.

archibate/dotfiles-opencode108Updated Apr 29, 2026

Interact with local Chrome browser session (only on explicit user approval after being asked to inspect, debug, or interact with a page open in Chrome)

The English summary is being prepared. Showing the original description.

archibate/dotfiles-opencode108Updated Apr 29, 2026

See all skills by archibate

Report a problem with this skill