translation-quality-defect-families

Classify recurring translation-quality defects into reusable correction families.

248|23|Updated May 10, 2026
One-click install
npx skills add https://github.com/SaberOnGo/public-domain-books-translation --skill translation-quality-defect-families
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: translation-quality-defect-families
Source: https://github.com/SaberOnGo/public-domain-books-translation/tree/main/skills/translation-quality-defect-families
Command: npx skills add https://github.com/SaberOnGo/public-domain-books-translation --skill translation-quality-defect-families

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill turns recurring translation-quality problems into reusable correction patterns so the same issue can be found, fixed, and prevented across a whole book project.

Core Features & Use Cases

  • Defect family classification: Groups repeated fidelity, fluency, terminology, rhythm, and readability problems into one actionable pattern.
  • Low-token auditing: Supports targeted scans with source comparison, exact-string searching, glossary checks, and spot-check review instead of full rereads.
  • Reusable lesson backfill: Records the finding method, risk, fix pattern, and exceptions so later reviewers do not rediscover the same issue.
  • Use case: A translation review finds repeated stiff phrasing or terminology drift in several chapters, and this Skill helps audit similar passages, fix confirmed matches, and document the family rule in the same round.

Quick Start

Use the translation-quality-defect-families skill to classify the recurring issue, scan for similar passages, fix confirmed matches, and record the reusable lesson with a fresh recheck.

Frequently Asked Questions about translation-quality-defect-families

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I catch recurring translation defects across a whole book project?

You can catch recurring translation defects by classifying repeated fidelity, fluency, terminology, and rhythm issues into reusable correction families, then scanning similar passages to fix confirmed matches in the same review round.

What is the best way to audit terminology drift without a full reread?

The best way to audit terminology drift without a full reread is using low-token source comparison, exact-string or regex scanning, and targeted glossary checks to spot-check passages and identify deviations.

How do I track translation quality fixes and exceptions during spot checks?

Track translation quality fixes by recording the finding method, risk, fix pattern, and justified exceptions for each defect family, enabling reusable lesson backfill so later reviewers avoid rediscovering the same issue.

Can I use this for post-EPUB spot-check workflows?

Yes, this workflow supports post-EPUB spot-checks by applying fidelity, fluency, terminology, rhythm, and style-debt auditing through targeted source comparison, exact-string scanning, and same-round recheck backfill.