Blog Translate, SEO-Optimized Blog Translation
Translates an existing blog post into one or more target languages. Unlike
generic translation, this skill produces SEO-optimized, publication-ready
content with localized keywords, meta tags, and culturally correct
formatting.
Adapted from claude-blog-multilingual by Chris Mueller (Pro Hub Challenge,
March 2026). Original: https://github.com/Chriss54/multilingual-int
Key References
Load on demand:
references/translation-rules.md, format preservation, number/date/currency
formats per locale, quote handling, quality criteria.
references/cultural-adaptation.md, cultural profiles per locale (DACH,
Francophone, Hispanic, Japanese, custom). This file is shared with
blog-localize (do not duplicate).
Workflow
Phase 1: Input Parsing
- Read the source file (markdown, MDX, or HTML) only after resolving it
against the project root/current working directory. Reject symlinked paths,
traversal outside the root, files over 10 MB, and binary files.
- Auto-detect source language. Order of preference:
- Frontmatter
lang field.
- HTML
lang attribute.
- Content analysis (script, common stop words).
- Parse target languages from
--to as comma-separated Google-compatible
hreflang tags (de, fr, es-MX, ja, pt-BR, zh-Hant).
If --to is missing, ask the user once: "Which languages should I
translate to? Provide hreflang tags such as de, fr, es-MX, ja, pt-BR."
- Normalize every code with the shared multilingual locale rules used by
blog-multilingual, blog-localize, and blog-locale-audit: ISO 639-1
language in lowercase, optional ISO 15924 script in title case, optional
ISO 3166-1 Alpha-2 region in uppercase. Reject invalid codes with a
suggestion (jp becomes "Did you mean ja for Japanese?"). Require a
region or explicit neutral mode for ambiguous language-only targets such as
es, pt, and zh. If a target equals the normalized source language,
skip it with a notice.
Phase 2: Content Analysis
Extract the translatable surface:
- Frontmatter:
title, description, tags, author (only when
translatable, e.g. role labels, not personal names).
- All headings (H1, H2, H3).
- Body paragraphs.
- Image
alt text and <figcaption> content.
- Chart
<text> and <tspan> content; preserve every SVG attribute (x,
y, font-size, fill, transform).
- FAQ questions and answers.
- Evidence-backed explanation text.
- Key Takeaways or summary box.
- CTA text.
- Internal-link zone anchor text.
Preserve unchanged:
- Markdown and HTML structure, tags, attributes.
- Image URLs, link URLs, frontmatter keys.
- Executable code fences and inline code. Translate comments only outside
executable code, or when the user explicitly asks for localized tutorial
comments.
- Internal-link zone markers (
[INTERNAL-LINK: ...]).
- Source organization names in citations (Gartner, McKinsey, etc.).
- Person names.
- Schema JSON-LD blocks (translate user-facing content strings only; never
translate Person, Organization, or Brand names, URLs, IDs,
@id, or
sameAs).
Identify the primary and secondary keywords for Phase 3.
Phase 3: Keyword Localization
For each target language:
- Decide whether the source keyword is the established term in the target
market. If yes (e.g., "Content Marketing" stays in German), keep it.
- If a local equivalent has real search behavior, swap to it.
- Apply the same logic to secondary keywords.
- Record the mapping. The translator agent uses it to update title, meta
description, and H2 headings consistently.
Phase 4: Translation
Spawn the blog-translator agent (via Task) for each target language with:
- The source content.
- The keyword localization map from Phase 3.
- The target language code.
- Pointer to
references/translation-rules.md.
- Instruction to limit this phase to language, register, natural topical
coverage, and formatting. Reserve brand, legal, statistic, and cultural
substitutions for
blog-localize.
Run agents in parallel when translating into multiple languages.
The agent returns the fully translated post in the same format as the input.
Phase 5: Post-Processing
For each translated version:
- Add or update locale frontmatter:
lang: "de"
translatedFrom: "en"
translatedDate: "YYYY-MM-DD"
slug: "wie-man-ki-slop-vermeidet"
- Verify structural integrity:
- Same number of H2 and H3 sections as the original.
- All images present with translated alt text.
- All SVG charts present with translated text labels (length-adjusted:
DE +30%, FR +15%, JA -20%, others see
references/translation-rules.md).
- FAQ count matches.
- Evidence-backed explanations from the source remain intact where present.
- Save translated files:
translations/
{lang}/{localized-slug}.{ext}
When invoked from blog-multilingual, save into
multilingual/{lang}/{localized-slug}.{ext} instead.
Before writing, slugify {localized-slug} to lowercase ASCII with only
a-z, 0-9, and hyphens, rejecting empty or reserved names. Create the
language directory from the normalized hreflang code only. Resolve the
final path and require it to stay inside the intended output root; reject
symlinked output paths.
Phase 6: Translation-Quality Guardrails
Scan the output for machine-translation artifacts before reporting done:
- Literal idioms (English idioms transliterated, not adapted).
- Unnatural word order (SOV translated as SVO into a non-SVO language, or
the reverse).
- Mixed-language sentences (other than established loanwords).
- Number, date, or currency strings still in source format.
- Frontmatter strings still in the source language.
Flag every issue inline (file path, line number, fix suggestion). The
translator agent should re-pass any flagged passage before delivery.
Phase 7: Delivery
## Translation complete: [Original title]
### Source
- Language: [source]
- File: [source path]
### Translations
| Language | File | Keywords adapted | Status |
|----------|------|------------------|--------|
| de | translations/de/{slug}.md | [N] | ok |
| fr | translations/fr/{slug}.md | [N] | ok |
### Quality checks
- Structural integrity: pass / fail per language
- Meta tags localized: pass / fail per language
- Numbers, dates, currencies formatted per locale: pass / fail
- Keywords localized: [N] keywords adapted
- Machine-translation artifacts flagged: [N] (see notes above)
### Next steps
- Run `/blog localize <file> --locale <code>` for cultural deep-adaptation.
- Run `/blog locale-audit translations/` to verify completeness.
- Use `/blog multilingual` to combine write, translate, localize, hreflang
in one command.
Error Handling
| Scenario |
Action |
| Unsupported language code |
Suggest the correct Google-compatible hreflang code |
| Source equals a target |
Skip with "Source is already in [lang]" |
| File not found |
Report error with suggested path |
| Translator agent timeout |
Retry once, then report partial results |
| Binary or non-text file |
Report error, suggest correct file |
Cross-References
- Next step (cultural depth):
/blog localize <file> --locale <code>
- QA sweep across all language versions:
/blog locale-audit <directory>
- One-command pipeline:
/blog multilingual <topic> --languages <codes>
1---
2name: blog-translate
3description: >
4 Translate existing blog posts into one or more target languages with
5 SEO-optimized localization. Produces native-quality translations that
6 preserve markdown structure, frontmatter, schema JSON-LD, image and chart
7 embeds, and evidence-backed explanations. Localizes keywords, meta tags, numbers,
8 dates, currencies, and quote styles per locale. Flags machine-translation
9 artifacts for review. Run BEFORE blog-localize: this handles language
10 conversion; localize handles cultural adaptation after translation
11 completes.
12 Use when user says "translate blog", "blog translate", "uebersetzen",
13 "traduire", "traducir", "translate post", "blog auf Deutsch", "blog en
14 espanol".
15user-invokable: true
16argument-hint: "<file> --to <comma-separated-codes>"
17license: MIT
18compatibility: Standalone within claude-blog. Invoked by blog-multilingual.
19metadata:
20 author: AgriciDaniel
21 version: "2.2.0"
22 category: blog
23---
24
25# Blog Translate, SEO-Optimized Blog Translation
26
27Translates an existing blog post into one or more target languages. Unlike
28generic translation, this skill produces SEO-optimized, publication-ready
29content with localized keywords, meta tags, and culturally correct
30formatting.
31
32> Adapted from `claude-blog-multilingual` by Chris Mueller (Pro Hub Challenge,
33> March 2026). Original: https://github.com/Chriss54/multilingual-int
34
35## Key References
36
37Load on demand:
38
39- `references/translation-rules.md`, format preservation, number/date/currency
40 formats per locale, quote handling, quality criteria.
41- `references/cultural-adaptation.md`, cultural profiles per locale (DACH,
42 Francophone, Hispanic, Japanese, custom). This file is shared with
43 `blog-localize` (do not duplicate).
44
45## Workflow
46
47### Phase 1: Input Parsing
48
491. Read the source file (markdown, MDX, or HTML) only after resolving it
50 against the project root/current working directory. Reject symlinked paths,
51 traversal outside the root, files over 10 MB, and binary files.
522. Auto-detect source language. Order of preference:
53 - Frontmatter `lang` field.
54 - HTML `lang` attribute.
55 - Content analysis (script, common stop words).
563. Parse target languages from `--to` as comma-separated Google-compatible
57 hreflang tags (`de`, `fr`, `es-MX`, `ja`, `pt-BR`, `zh-Hant`).
58 If `--to` is missing, ask the user once: "Which languages should I
59 translate to? Provide hreflang tags such as de, fr, es-MX, ja, pt-BR."
604. Normalize every code with the shared multilingual locale rules used by
61 `blog-multilingual`, `blog-localize`, and `blog-locale-audit`: ISO 639-1
62 language in lowercase, optional ISO 15924 script in title case, optional
63 ISO 3166-1 Alpha-2 region in uppercase. Reject invalid codes with a
64 suggestion (`jp` becomes "Did you mean `ja` for Japanese?"). Require a
65 region or explicit neutral mode for ambiguous language-only targets such as
66 `es`, `pt`, and `zh`. If a target equals the normalized source language,
67 skip it with a notice.
68
69### Phase 2: Content Analysis
70
71Extract the translatable surface:
72
73- Frontmatter: `title`, `description`, `tags`, `author` (only when
74 translatable, e.g. role labels, not personal names).
75- All headings (H1, H2, H3).
76- Body paragraphs.
77- Image `alt` text and `<figcaption>` content.
78- Chart `<text>` and `<tspan>` content; preserve every SVG attribute (`x`,
79 `y`, `font-size`, `fill`, `transform`).
80- FAQ questions and answers.
81- Evidence-backed explanation text.
82- Key Takeaways or summary box.
83- CTA text.
84- Internal-link zone anchor text.
85
86Preserve unchanged:
87
88- Markdown and HTML structure, tags, attributes.
89- Image URLs, link URLs, frontmatter keys.
90- Executable code fences and inline code. Translate comments only outside
91 executable code, or when the user explicitly asks for localized tutorial
92 comments.
93- Internal-link zone markers (`[INTERNAL-LINK: ...]`).
94- Source organization names in citations (Gartner, McKinsey, etc.).
95- Person names.
96- Schema JSON-LD blocks (translate user-facing content strings only; never
97 translate Person, Organization, or Brand names, URLs, IDs, `@id`, or
98 `sameAs`).
99
100Identify the primary and secondary keywords for Phase 3.
101
102### Phase 3: Keyword Localization
103
104For each target language:
105
1061. Decide whether the source keyword is the established term in the target
107 market. If yes (e.g., "Content Marketing" stays in German), keep it.
1082. If a local equivalent has real search behavior, swap to it.
1093. Apply the same logic to secondary keywords.
1104. Record the mapping. The translator agent uses it to update title, meta
111 description, and H2 headings consistently.
112
113### Phase 4: Translation
114
115Spawn the `blog-translator` agent (via Task) for each target language with:
116
117- The source content.
118- The keyword localization map from Phase 3.
119- The target language code.
120- Pointer to `references/translation-rules.md`.
121- Instruction to limit this phase to language, register, natural topical
122 coverage, and formatting. Reserve brand, legal, statistic, and cultural
123 substitutions for `blog-localize`.
124
125Run agents in parallel when translating into multiple languages.
126
127The agent returns the fully translated post in the same format as the input.
128
129### Phase 5: Post-Processing
130
131For each translated version:
132
1331. Add or update locale frontmatter:
134 ```yaml
135 lang: "de"
136 translatedFrom: "en"
137 translatedDate: "YYYY-MM-DD"
138 slug: "wie-man-ki-slop-vermeidet"
139 ```
1402. Verify structural integrity:
141 - Same number of H2 and H3 sections as the original.
142 - All images present with translated alt text.
143 - All SVG charts present with translated text labels (length-adjusted:
144 DE +30%, FR +15%, JA -20%, others see `references/translation-rules.md`).
145 - FAQ count matches.
146 - Evidence-backed explanations from the source remain intact where present.
1473. Save translated files:
148 ```
149 translations/
150 {lang}/{localized-slug}.{ext}
151 ```
152 When invoked from `blog-multilingual`, save into
153 `multilingual/{lang}/{localized-slug}.{ext}` instead.
154 Before writing, slugify `{localized-slug}` to lowercase ASCII with only
155 `a-z`, `0-9`, and hyphens, rejecting empty or reserved names. Create the
156 language directory from the normalized hreflang code only. Resolve the
157 final path and require it to stay inside the intended output root; reject
158 symlinked output paths.
159
160### Phase 6: Translation-Quality Guardrails
161
162Scan the output for machine-translation artifacts before reporting done:
163
164- Literal idioms (English idioms transliterated, not adapted).
165- Unnatural word order (SOV translated as SVO into a non-SVO language, or
166 the reverse).
167- Mixed-language sentences (other than established loanwords).
168- Number, date, or currency strings still in source format.
169- Frontmatter strings still in the source language.
170
171Flag every issue inline (file path, line number, fix suggestion). The
172translator agent should re-pass any flagged passage before delivery.
173
174### Phase 7: Delivery
175
176```
177## Translation complete: [Original title]
178
179### Source
180- Language: [source]
181- File: [source path]
182
183### Translations
184| Language | File | Keywords adapted | Status |
185|----------|------|------------------|--------|
186| de | translations/de/{slug}.md | [N] | ok |
187| fr | translations/fr/{slug}.md | [N] | ok |
188
189### Quality checks
190- Structural integrity: pass / fail per language
191- Meta tags localized: pass / fail per language
192- Numbers, dates, currencies formatted per locale: pass / fail
193- Keywords localized: [N] keywords adapted
194- Machine-translation artifacts flagged: [N] (see notes above)
195
196### Next steps
197- Run `/blog localize <file> --locale <code>` for cultural deep-adaptation.
198- Run `/blog locale-audit translations/` to verify completeness.
199- Use `/blog multilingual` to combine write, translate, localize, hreflang
200 in one command.
201```
202
203## Error Handling
204
205| Scenario | Action |
206|----------|--------|
207| Unsupported language code | Suggest the correct Google-compatible hreflang code |
208| Source equals a target | Skip with "Source is already in [lang]" |
209| File not found | Report error with suggested path |
210| Translator agent timeout | Retry once, then report partial results |
211| Binary or non-text file | Report error, suggest correct file |
212
213## Cross-References
214
215- Next step (cultural depth): `/blog localize <file> --locale <code>`
216- QA sweep across all language versions: `/blog locale-audit <directory>`
217- One-command pipeline: `/blog multilingual <topic> --languages <codes>`