Install
$ agentstack add skill-lovstudio-skills-translation-review ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
translation-review — Chinese-English Translation Review
Systematically compare a Chinese source document against its English translation, identify issues across 6 dimensions, and produce a prioritized review report.
When to Use
- User provides a Chinese original + English translation pair for review
- User asks to check/proofread an English translation from Chinese
- User mentions "翻译检查", "审校", "translation review"
Workflow (MANDATORY)
You MUST follow these steps in order:
Step 1: Collect Input
Use AskUserQuestion to confirm before starting:
Ask the user:
- Document type — academic paper (SCI/EI), technical docs, business docs, general
- Domain — medicine, CS, law, finance, general (affects terminology checking)
- Focus areas — terminology accuracy, grammar, style, references, all (multi-select)
If the user has already provided context (e.g., "SCI 4区医学论文"), skip the obvious questions.
Step 2: Read Both Documents
Read the Chinese source and English translation. For .docx files:
pandoc ".docx" -t plain --wrap=none
pandoc ".docx" -t plain --wrap=none
For .md or .txt files, read directly.
If files are large, read in sections (offset/limit) to stay within context.
Step 3: Systematic Review
Compare the two documents across 6 dimensions, in this order:
D1: Mistranslation (meaning changed)
- Words/phrases translated to a different meaning
- Omissions (content in Chinese missing from English)
- Additions (content in English not in Chinese)
- Pay special attention to domain-specific terms that have been incorrectly generalized or specialized
D2: Terminology Accuracy
- Domain-specific terms: check against standard English terminology
- Abbreviation consistency: first occurrence should have full form + abbreviation
- Proper nouns: names, institutions, gene/protein names, drug names
- Units and measurements
D3: Grammar & Syntax
- Subject-verb agreement
- Tense consistency (especially in Methods vs Results)
- Article usage (a/an/the) — common error in Chinese-to-English
- Dangling modifiers, run-on sentences
- Passive voice overuse
D4: Consistency
- Same Chinese term translated differently in different places
- Formatting inconsistencies (capitalization, hyphenation, number style)
- Heading style consistency
D5: References & Citations (academic papers only)
- Citation numbers match between Chinese and English versions
- Reference format consistency (author names, journal abbreviations)
- Missing author information in reference entries
- In-text citation correctness
D6: Style & Register
- Academic register appropriate for target journal
- Overly verbose or literary phrasing
- Chinese-influenced sentence structure (e.g., "through...method" patterns)
- Paragraph length and structure
Step 4: Source Document Issues
While reviewing, also note issues in the Chinese source that the English faithfully translates but are themselves incorrect. Flag these separately.
Step 5: Generate Report
Output a structured review report with the following format:
# 翻译审校报告 / Translation Review Report
**Document**: [title]
**Reviewer**: 手工川 (AI-assisted)
**Date**: [YYYY-MM-DD]
**Domain**: [domain]
**Document type**: [type]
---
## Summary
[2-3 sentence overall assessment]
---
## A-Level Issues (Must Fix)
### A1. [Issue title]
**Location**: [section/paragraph]
**Original (CN)**: "..."
**Translation (EN)**: "..."
**Problem**: [explanation]
**Suggested fix**: → "..."
---
## B-Level Issues (Recommended)
### B1. [Issue title]
...
---
## Source Document Issues
### C1. [Issue title]
...
---
## Priority Summary
| Priority | # | Issue | Impact |
|----------|---|-------|--------|
| Must fix | A1 | ... | ... |
| ... | | | |
---
## Conclusion
[Overall quality assessment + recommended next steps]
Step 6: Save Report
Save the report to the same directory as the input files:
Filename: 手工川-translation-review-{YYYY-MM-DD}-v0.1.md
If a file with this name exists, increment the version.
Issue Severity Classification
| Level | Criteria | Examples | |-------|----------|---------| | A (Must fix) | Meaning changed, factual error, or would cause rejection | Mistranslation, wrong terminology, reference number mismatch | | B (Recommended) | Awkward but understandable, or non-standard usage | Verbose phrasing, minor grammar, style inconsistency | | C (Source issue) | Error in Chinese original, faithfully translated | Wrong terminology in Chinese, factual inaccuracy |
Domain-Specific Checklists
Academic / SCI Papers
- [ ] Abstract: structured, matches body content
- [ ] Keywords: standard MeSH or domain terms
- [ ] Reference numbers: sequential, match between CN/EN versions
- [ ] Figure/table captions: translated and consistent
- [ ] Statistical reporting: p-values, CI format
Medical / Biomedical
- [ ] Disease names: ICD/MeSH standard
- [ ] Gene/protein names: HUGO/UniProt standard (italics for genes)
- [ ] Drug names: INN (International Nonproprietary Names)
- [ ] Anatomical terms: Terminologia Anatomica standard
Technical Documentation
- [ ] UI strings: consistent with product localization
- [ ] Code snippets: untranslated, syntax preserved
- [ ] API terms: match official documentation
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: lovstudio
- Source: lovstudio/skills
- License: MIT
- Homepage: https://lovstudio.ai/skills
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.