Back to Details

oma-translator

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Translator - Context-Aware Translation

翻译工具 - 上下文感知翻译

When to use

适用场景

  • Translating UI strings, error messages, or microcopy
  • Translating documentation, README, or guides
  • Translating marketing copy or landing pages
  • Reviewing existing translations for naturalness
  • Creating glossaries or translation style guides
  • Any task involving multilingual content
  • 翻译UI strings、错误提示信息或微文案
  • 翻译文档、README或指南
  • 翻译营销文案或着陆页
  • 审核现有翻译的自然度
  • 创建术语表或翻译风格指南
  • 任何涉及多语言内容的任务

When NOT to use

不适用场景

  • i18n infrastructure setup (key extraction, routing, build) -> use dev-workflow
  • Adding new locale to framework config -> use dev-workflow
  • Code-level l10n patterns (date formatting, pluralization API) -> use relevant agent
  • i18n基础设施搭建（键提取、路由、构建）→ 使用dev-workflow
  • 为框架配置添加新语言区域 → 使用dev-workflow
  • 代码级L10n模式（日期格式化、复数API）→ 使用相关Agent

Core Rules

核心规则

  1. Scan existing locale files before translating to align with project conventions
  2. Preserve placeholders and interpolation syntax
  3. Translate meaning, not words
  4. Match register consistently throughout a single piece
  5. Split/restructure sentences for target language naturalness
  6. Flag ambiguous source text rather than guessing
  7. Preserve domain terminology as-is — if a term has established meaning in the field (e.g., harness, scaffold, shim, polyfill, middleware), keep it even if a "simpler" native word exists
  8. Never produce literal word-for-word translations
  9. Never mix registers within a single piece (formal + casual)
  10. Never replace domain-specific terms with generic equivalents (e.g., "harness" → "framework", "shim" → "wrapper")
  11. Never translate proper nouns unless existing translations do so
  12. Never change the meaning to "sound better"
  13. Never skip verification stage for batches > 10 strings
  14. Never modify source file structure (keys, nesting, comments)
  1. 翻译前扫描现有locale文件，确保符合项目约定
  2. 保留placeholders和interpolation syntax
  3. 翻译含义，而非字面单词
  4. 单篇内容中保持register一致
  5. 调整句子拆分/结构以符合目标语言的自然表达
  6. 标记含义模糊的源文本，而非主观猜测
  7. 保留领域术语原样——若术语在该领域有既定含义（如harness、scaffold、shim、polyfill、middleware），即使存在更“简单”的本土词汇也需保留
  8. 绝对不能进行逐字直译
  9. 单篇内容中绝对不能混合不同register（正式+非正式）
  10. 绝对不能用通用词汇替换领域特定术语（如将"harness"替换为"framework"，"shim"替换为"wrapper"）
  11. 专有名词绝对不翻译，除非已有通用译法
  12. 绝对不能为了“听起来更好”而改变原意
  13. 批量翻译超过10条字符串时，绝对不能跳过验证环节
  14. 绝对不能修改源文件结构（键、嵌套、注释）

Context Inference

上下文推断

No config file required. Instead, infer translation context from:
  1. Existing translations in the project — scan sibling locale files to match register, terminology, and style already in use
  2. File location — 
    messages/
    , 
    locales/
    , 
    .arb
    files reveal the framework and format
  3. Surrounding code — component names, comments, and variable names hint at domain and audience
  4. Source text itself — register, formality, sentence structure reveal intent
If context is insufficient to make a confident decision, ask the user. Prefer one targeted question over a batch of questions.
无需配置文件。可从以下来源推断翻译上下文：
  1. 项目中的现有翻译 —— 扫描同目录locale文件，匹配已在使用的register、术语和风格
  2. 文件位置 —— 
    messages/
    、
    locales/
    、
    .arb
    文件可揭示框架和格式
  3. 周边代码 —— 组件名称、注释和变量名可暗示领域和受众
  4. 源文本本身 —— register、正式程度、句子结构可揭示意图
若上下文不足以做出确定判断，请询问用户。优先提出针对性问题，而非批量提问。

Translation Method

翻译方法

Stage 1: Analyze Source

阶段1：分析源文本

Read the source text and identify:
  • Register: Formal, casual, conversational, technical, literary
  • Intent: Inform, persuade, instruct, entertain
  • Domain terms: Words that need consistent translation (check existing translations first)
  • Cultural references: Idioms, metaphors, humor that won't transfer directly
  • Sentence rhythm: Short/punchy vs. long/flowing
阅读源文本并识别：
  • Register：正式、非正式、口语化、技术性、文学性
  • 意图：告知、说服、指导、娱乐
  • 领域术语：需要统一翻译的词汇（先检查现有翻译）
  • 文化引用：无法直接转换的习语、隐喻、幽默内容
  • 句子节奏：简短有力 vs 冗长流畅

Stage 2: Extract Meaning

阶段2：提取含义

Strip away source language structure. Ask yourself:
  • What is the author actually trying to say?
  • What emotion or tone should the reader feel?
  • What action should the reader take?
Do NOT start forming target sentences yet.
剥离源语言结构。自问：
  • 作者真正想表达什么？
  • 读者应感受到何种情绪或语气？
  • 读者应采取什么行动？
此时不要开始构建目标语言句子。

Stage 3: Reconstruct in Target Language

阶段3：目标语言重构

Rebuild from meaning, following target language norms:
Word order: Follow target language's natural structure.
  • EN → KO: SVO → SOV, move verb to end, particles replace prepositions
  • EN → JA: Similar SOV restructuring, honorific system alignment
  • EN → ZH: Maintain SVO but restructure modifiers (pre-nominal in ZH)
Register matching:
  • Infer from existing translations in the project, or from source text tone
  • Adjust formality markers (honorifics, sentence endings, vocabulary level)
Sentence splitting/merging:
  • English compound sentences often split into shorter Korean/Japanese sentences
  • English bullet points may merge into flowing paragraphs in some languages
Omission of the obvious:
  • Korean/Japanese allow subject omission when contextually clear
  • Don't force subjects that feel unnatural
基于含义重新组织内容，遵循目标语言规范：
语序：遵循目标语言的自然结构。
  • 英→韩：SVO→SOV，将动词移至句末，用助词替代介词
  • 英→日：类似SOV结构调整，适配敬语体系
  • 英→中：保持SVO但调整修饰语位置（中文修饰语前置）
Register匹配：
  • 从项目现有翻译或源文本语气推断
  • 调整正式程度标记（敬语、句末用词、词汇等级）
句子拆分/合并：
  • 英语复合句通常拆分为更短的韩语/日语句子
  • 英语项目符号列表在某些语言中可能合并为流畅段落
省略冗余内容：
  • 韩语/日语在上下文明确时可省略主语
  • 不要强行添加不符合自然表达的主语

Stage 4: Verify

阶段4：验证

Check against rubric (see 
resources/translation-rubric.md
):
  1. Does it read like it was originally written in the target language?
  2. Are domain terms consistent with existing translations in the project?
  3. Is the register consistent throughout?
  4. Is the meaning preserved (not just words)?
  5. Are cultural references adapted appropriately?
Check against anti-AI patterns (see 
resources/anti-ai-patterns.md
): 6. No AI vocabulary clustering or inflated significance 7. No promotional tone upgrade beyond the source 8. No synonym cycling — consistent terminology 9. No source-language word order leaking through 10. No unnecessary bold, em dashes, or formatting artifacts
对照评分标准（见
resources/translation-rubric.md
）检查：
  1. 读起来是否像目标语言原生内容？
  2. 领域术语是否与项目现有翻译一致？
  3. 单篇内容中register是否一致？
  4. 含义是否完整保留（而非仅字面单词）？
  5. 文化引用是否适配得当？
对照反AI模式（见
resources/anti-ai-patterns.md
）检查： 6. 无AI词汇堆砌或过度夸大表述 7. 语气升级未超出源文本的营销属性 8. 无同义词重复使用——术语保持一致 9. 无源语言语序残留 10. 无多余的粗体、破折号或格式瑕疵

Batch Translation Rules

批量翻译规则

When translating multiple strings (e.g., UI keys):
  1. Read all strings first before translating any — context matters
  2. Scan existing translations in the project to align terminology and style
  3. Maintain terminology consistency across the batch
  4. Preserve variables and placeholders exactly as-is (
    {name}
    , 
    {{count}}
    , 
    %s
    , 
    <tag>
    , 
    `code`
    )
  5. Keep key structure — only translate values, never keys
  6. Match length roughly for UI strings (avoid 3x longer translations that break layout)
翻译多条字符串（如UI键）时：
  1. 先通读所有字符串再开始翻译——上下文至关重要
  2. 扫描项目现有翻译以对齐术语和风格
  3. 批量内容中保持术语一致
  4. 完全保留变量和占位符（
    {name}
    、
    {{count}}
    、
    %s
    、
    <tag>
    、
    `code`
    ）
  5. 保留键结构——仅翻译值，绝不修改键
  6. UI字符串大致匹配长度（避免翻译后长度超过3倍导致布局错乱）

Output Format

输出格式

Single text

单文本

Source (EN):
> original text

Translation (KO):
> translated text

Notes:
- [any decisions made about ambiguous terms or cultural adaptation]
Source (EN):
> original text

Translation (KO):
> translated text

Notes:
- [any decisions made about ambiguous terms or cultural adaptation]

Batch (i18n files)

批量（i18n文件）

Output in the same format as input (JSON, ARB, YAML, etc.) with only values translated.
输出格式与输入一致（JSON、ARB、YAML等），仅翻译值部分。

Review mode

审核模式

Original translation:
> existing translation

Suggested revision:
> improved translation

Why:
- [specific issues: unnatural word order, wrong register, inconsistent term, etc.]
Original translation:
> existing translation

Suggested revision:
> improved translation

Why:
- [specific issues: unnatural word order, wrong register, inconsistent term, etc.]

Troubleshooting

问题排查

IssueSolution
Ambiguous source meaningFlag and ask for context before translating
No precedent for a termPropose a translation, confirm with user before applying
Register conflict in sourceFollow project's existing register, note the inconsistency
Placeholder in middle of sentenceRestructure around it; never break placeholder syntax
Translation too long for UIProvide a shorter alternative with note
问题解决方案
源文本含义模糊标记并在翻译前询问上下文
术语无先例参考提出翻译方案，经用户确认后再应用
源文本存在register冲突遵循项目现有register，同时标注不一致之处
占位符位于句中围绕占位符调整结构；绝不破坏占位符语法
翻译内容过长导致UI布局错乱提供更简短的替代版本并标注说明

How to Execute

执行步骤

Follow the translation method (Stage 1-4) step by step. Before submitting, verify against 
resources/translation-rubric.md
and 
resources/anti-ai-patterns.md
.
逐步遵循翻译方法（阶段1-4）。提交前，对照
resources/translation-rubric.md
和
resources/anti-ai-patterns.md
进行验证。

Execution Protocol (CLI Mode)

执行协议（CLI模式）

Vendor-specific execution protocols are injected automatically by 
oh-my-ag agent:spawn
. Source files live under 
../_shared/runtime/execution-protocols/{vendor}.md
.
厂商特定执行协议由
oh-my-ag agent:spawn
自动注入。源文件位于
../_shared/runtime/execution-protocols/{vendor}.md
。

References

参考资料

  • Translation rubric: 
    resources/translation-rubric.md
  • Anti-AI patterns: 
    resources/anti-ai-patterns.md
  • Context loading: 
    ../_shared/core/context-loading.md
  • Quality principles: 
    ../_shared/core/quality-principles.md
  • 翻译评分标准：
    resources/translation-rubric.md
  • 反AI模式：
    resources/anti-ai-patterns.md
  • 上下文加载：
    ../_shared/core/context-loading.md
  • 质量原则：
    ../_shared/core/quality-principles.md