Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ur-grue/autopunk-media-skills --skill translation-accuracy-reviewergit clone --depth 1 https://github.com/ur-grue/autopunk-media-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ur-grue/autopunk-media-skills/translation-accuracy-reviewer)<a href="https://agentmods.dev/skills/ur-grue/autopunk-media-skills/translation-accuracy-reviewer"><img src="https://agentmods.dev/badge/skills/ur-grue/autopunk-media-skills/translation-accuracy-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ur-grue/autopunk-media-skills/translation-accuracy-reviewer"><img src="https://agentmods.dev/badge/skills/ur-grue/autopunk-media-skills/translation-accuracy-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00036 | $0.01748 |
| Opus 5 | $0.00018 | $0.00874 |
| Sonnet 5 | $0.00007 | $0.00350 |
| Haiku 4.5 | $0.00004 | $0.00175 |
Grade A, and why
translation-accuracy-reviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Translation Accuracy Reviewer
What This Skill Does
Compares a source text against its translation, flags potential inaccuracies and mistranslations, and explains each issue so an editor can make an informed correction.
When To Use This Skill
- Before publishing a translated article, press release, interview transcript, or official statement
- When a machine translation (DeepL, Google Translate) has been used as a first draft and needs human-quality review
- When a freelance translation arrives and the editor is not fluent in the source language
- When a quoted source's words have been translated and accuracy is legally or editorially critical (court testimony, official denials, diplomatic statements)
What You Need To Provide
Required: The original source text (any language) and the translated version side by side, with the source and target languages clearly labelled. Optional: The publication context (news article, documentary subtitle, legal document, press release); the intended audience for the translation; any known sensitivities (e.g., "the source's phrasing on this point is politically significant"); a note on whether the translation should be literal or idiomatic.
How the Assistant Approaches This
- Reads the source text in full to understand its meaning, tone, register, and any culturally specific references or idioms before looking at the translation.
- Works through the translation sentence by sentence against the source, flagging: omissions (content present in the source that is missing from the translation), additions (content in the translation with no basis in the source), mistranslations (words or phrases rendered inaccurately), register mismatches (source is formal, translation is casual, or vice versa), and cultural or idiomatic equivalence issues where a literal rendering distorts the intended meaning.
- Produces a structured report: each flag is numbered, the source passage and translated passage are quoted together, the problem is named clearly in plain language, and a more accurate alternative is provided. Closes with an overall accuracy assessment.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 121 lines · 36 tokens per session scan A dcf65c86345d
translation-accuracy-reviewer is a skill published in the GitHub repository ur-grue/autopunk-media-skills (32 stars, last pushed 12d ago), licensed MIT. It adds 36 tokens to every session and 1,748 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
youtube-transcribe
A tool that downloads YouTube videos or subtitles, turns speech into text, and can translate English into Chinese in a bilingual Markdown file.
localization-engineer
Expert in internationalization (i18n), multi-language support, and localization.
nextjs-i18n
Best practices for multi-language handling, locale routing, and detection strategies across App and Pages Router. Use when adding i18n, locale routing, or language detection in Next.js.
paper-translation
Use when translating academic papers (PDF to Chinese Markdown), extracting PDFs with MinerU API, or needing to preserve LaTeX formulas/images/tables through a translation workflow.
cf-translate
Translate publication-ready ContentForge content into any of 15 languages at three localization levels (literal, adapted, transcreated) — preserving brand voice, keeping citation URLs and DOIs untouched, and adapting SEO keyword placements for the target market — delivered as a translated .docx plus a quality report…
russian-editorial-review
A Russian-language editorial checklist based on the Milchin and Cheltsova publishing reference. It organizes a full review of names, capitalization, abbreviations, numbers, dates, lists, quotations, bibliographies, and typography.