Kostja Zhang

translation

Contentv1.0.1
By Kostja ZhangUpdated 2026-05-112 versions · latest released 2026-03-30@clawhub_kostja94/translation
Rewriting

According to AlphaMoat, as of 2026-09, translation has 4,901 downloads and 0 stars — ranked #1,901 of 5,078 AI skills overall, and #144 of 362 in Content.

Installs
103
Downloads
4.9K
Stars
-
Overall Rank
#1,901
Top 38% overall · Top 40% in Content
PlatformInstallsDownloadsStarsComments
ClawHub ↗1034.9K--
SkillHub ↗964.0K-—

Platforms count differently and update at different times.

What this skill does

It covers audience and tone briefs, glossaries, style guides, when to use human or machine translation, translation memory, quality review, and multilingual SEO. The document recommends human review for customer-facing or brand-critical content and distinguishes drafts from final copy; terminology and compliance requirements should still be checked by professionals familiar with the target market.

Profile published 2026-10-29

What you can do with it

  • Prepare project context, audience, tone, and terminology requirements for a translation.
  • Establish glossaries and style rules for consistency across languages.
  • Plan machine-translation post-editing, human review, and multilingual SEO checks.

When to use it

  • When translating marketing, product, technical, or compliance content.
  • When keeping terminology and brand voice consistent across locales.
  • Before publishing a translation, when a native speaker or target-market specialist should review it.

Example prompts

Generated from the skill’s documentation as examples; not results of actual runs.

Draft a translator brief and glossary structure from this material without translating or publishing content.
List the terminology, tone, and length constraints that need human review after machine-translating this customer-facing UI text.

Monthly Trend

Downloads4.9K
6652.9K5.2K2026-03 · Downloads 9792026-05 · Downloads 2.5K2026-06 · Downloads 2.9K2026-07 · Downloads 3.6K2026-08 · Downloads 4.4K2026-09 · Downloads 4.9K26-0326-09
Stars-
0112026-03 · Stars -2026-05 · Stars -2026-06 · Stars -2026-07 · Stars -2026-08 · Stars -2026-09 · Stars -26-0326-09

The counting method changed in 2026-09: skills with the same name are now counted separately by author, and the full catalog is covered. Dashed lines show months before the change; they are not compared with later months, and no growth is computed across the change.

MonthDownloadsMoMStars
2026-094.9K—-
2026-084.4K+21.4%-
2026-073.6K+26.7%-
2026-062.9K+15.8%-
2026-052.5K—-
2026-03979—-

SKILL.md (original)

The author’s original file, shown as plain text without edits or review by AlphaMoat. Links are not clickable; check commands before running them. Fetched 2026-10-23.

8.5 KB · ClawHub v1.0.1
---
name: translation
description: When the user wants to translate content, create translation workflows, manage terminology, or optimize translation quality. Also use when the user mentions "translate," "translation," "localization copy," "glossary," "terminology," "style guide translation," "machine translation," "human translation," "TMS," or "multilingual content." For strategy, use localization-strategy.
metadata:
  version: 1.0.1
---

# Content: Translation

Guides translation workflow, terminology, style, and quality for multilingual content. Covers when to use human vs machine translation, glossary and style guide creation, and SEO considerations. For i18n implementation, hreflang, and URL structure, see **localization-strategy**.
Show all (223 lines)

**When invoking**: On **first use**, if helpful, open with 1–2 sentences on what this skill covers and why it matters, then provide the main output. On **subsequent use** or when the user asks to skip, go directly to the main output.

## Scope

- **Workflow**: Brief → translate → review
- **Terminology**: Glossary creation and management
- **Style guide**: Voice, tone, formatting per language
- **Human vs MT**: When to use each; post-editing
- **Quality**: QA, consistency, SEO
- **Market-specific**: Terminology by region

## Initial Assessment

**Check for project context first:** If `.claude/project-context.md` or `.cursor/project-context.md` exists, read it for brand voice, target markets, and product terminology.

Identify:
1. **Content type**: Product UI, marketing copy, blog, landing page, docs
2. **Target language(s)**: Priority locales
3. **Existing assets**: Glossary, style guide, translation memory (TM)

---

## 1. Translation Workflow

### Brief → Translate → Review

| Phase | Purpose | Output |
|-------|---------|--------|
| **Brief** | Context, audience, tone, glossary reference, style guide | Translator brief document |
| **Translate** | First pass; use glossary + TM when available | Draft translation |
| **Review** | Native speaker review; consistency, brand voice, SEO | Final translation |

### Translator Brief

Include in every project:
- **Context**: What the content is for (landing page, product UI, blog)
- **Audience**: Target market, user persona
- **Tone**: Formal, casual, technical
- **Glossary**: Link or attach; mandatory terms
- **Style guide**: Reference or key rules
- **Constraints**: Character limits (UI), SEO keywords to include naturally

### Content-Type Workflow

| Content Type | Approach | Notes |
|--------------|----------|-------|
| **Product UI** | Glossary-critical; short strings; consistency | Use TM; avoid machine translation |
| **Marketing copy** | Brand voice; cultural adaptation | Human translation; see terminology |
| **Blog / Article** | SEO; natural keyword placement | Re-research keywords in target language; don't translate keyword lists |
| **Landing page** | Conversion-focused; CTA clarity | Human; test localized CTAs |
| **Technical docs** | Precision; glossary | TM + glossary; consider MT + post-edit for high volume |

---

## 2. Glossary & Terminology

### Glossary Purpose

- **Consistency**: Same term translated the same way across all content
- **Brand**: Product names, feature names, approved phrasing
- **Compliance**: Safety, legal, regulated terms
- **Cost**: Reduces rework; accelerates review

### Glossary Structure

| Field | Purpose |
|-------|---------|
| **Source term** | Original (e.g., English) |
| **Target term** | Approved translation |
| **Context** | Where it appears; usage note |
| **Do not translate** | Brand names, product names (when applicable) |

### Market-Specific Terminology

| Term | English | Chinese | Notes |
|------|---------|---------|------|
| **Generative AI** | Generative AI, GenAI | AIGC (人工智能生成内容) | Use "AIGC" for China; "Generative AI" for English |
| **Influencer** | Influencer | KOL (关键意见领袖) | "KOL" common in Chinese marketing |
| **User** | User | 用户 | Context-dependent |
| **Dashboard** | Dashboard | 控制台, 仪表盘 | Choose one; document in glossary |

**Principle**: Don't translate terminology lists; research how target market searches and speaks. See **keyword-research** for multi-language keyword research.

---

## 3. Style Guide (Translation)

### Elements to Define

| Element | Purpose |
|---------|---------|
| **Voice** | Brand personality; formal vs casual |
| **Tone** | Varies by content type (support = helpful; marketing = persuasive) |
| **Register** | Formal (您) vs informal (你) in languages that distinguish |
| **Punctuation** | Quotation marks, spacing (e.g., no space before colon in French) |
| **Formatting** | Dates, numbers, units; locale-specific |
| **Forbidden** | Terms or phrases to avoid |

### Per-Language Considerations

- **Chinese**: Simplified vs Traditional; measure word usage
- **German**: Formal (Sie) vs informal (du); compound nouns
- **Japanese**: Honorifics; keigo for formal contexts
- **Arabic**: RTL; formal vs dialect

---

## 4. Human vs Machine Translation

### When to Use Human Translation

| Scenario | Reason |
|----------|--------|
| **Product UI** | Terminology, UX clarity, brand |
| **Marketing copy** | Persuasion, cultural nuance, CTAs |
| **Landing pages** | Conversion; tested copy |
| **Legal, compliance** | Accuracy, liability |
| **Brand-critical** | Taglines, campaign copy |

### When Machine Translation (MT) May Be Acceptable

| Scenario | Condition |
|----------|-----------|
| **High-volume, low-stakes** | Internal docs, user-generated content |
| **Draft / triage** | MT + human post-edit (MTPE) |
| **Real-time** | Chat, support; with disclaimer |

### Avoid

- **Raw MT for product/marketing**: Terminology errors, cultural misfires, poor SEO
- **MT without post-edit** for customer-facing content
- **Translating keyword lists** instead of re-researching in target language

---

## 5. Translation Memory (TM) & TMS

### Translation Memory

- **What**: Stores approved source↔target sentence pairs
- **Benefit**: Consistency; reuse; lower cost per word; faster turnaround
- **Best practice**: Maintain TM; clean duplicates; align with glossary

### Translation Management System (TMS)

- **Use for**: Centralized workflow; glossary + TM integration; vendor management
- **When**: Multiple languages; ongoing translation; team collaboration

---

## 6. Quality & SEO

### Quality Checklist

- [ ] Glossary terms used correctly
- [ ] Style guide followed
- [ ] No untranslated strings
- [ ] Numbers, dates, units localized
- [ ] Character limits respected (UI)
- [ ] Native speaker review completed

### SEO for Translated Content

- **Keywords**: Re-research in target language; don't translate from source
- **Metadata**: Title, description translated; see **title-tag**, **meta-description**
- **Hreflang**: Technical implementation in **localization-strategy**; translation produces the content
- **Thin content**: Avoid publishing many low-quality translated pages at once; can trigger penalties. See **localization-strategy** Multilingual Risks.

---

## 7. Integration with Localization

| Topic | Skill |
|-------|-------|
| **i18n implementation** | localization-strategy |
| **URL structure, hreflang** | localization-strategy |
| **Translation workflow, glossary, style** | This skill (translation) |
| **Keyword research by market** | keyword-research; localization-strategy |

When adding a new locale: create glossary, style guide, then translate. See **localization-strategy** for technical checklist (hreflang, sitemap, metadata).

---

## Output Format

- **Translator brief** (context, audience, glossary, style)
- **Glossary** additions or updates
- **Style guide** notes for target language
- **Human vs MT** recommendation
- **Quality** checklist

## Related Skills

### Strategy & Technical

- **localization-strategy**: i18n, hreflang, URL structure, pricing by market; translation produces content for localized pages
- **content-strategy**: Multilingual content planning; avoid thin translations
- **content-marketing**: Content types and formats; translation as one channel adaptation

### SEO & Content

- **keyword-research**: Multi-language keyword research; don't translate keyword lists
- **page-metadata**: Hreflang implementation
- **title-tag, meta-description**: Translate metadata per locale
- **copywriting**: Source copy to translate; brand voice
- **image-optimization**: Localize image filenames for translated pages

### Pages

- **article-page-generator**: Article structure; translate with SEO in mind
- **landing-page-generator**: Landing page copy; human translation for conversion

Popular in Content

Data month: 2026-09 · Downloads, stars and installs are aggregated monthly from public skill registries (ClawHub, SkillHub). See methodology.