renwei-zh
Use this skill when the user wants Chinese text reviewed, revised, or audited for formulaic model-like writing. The objective is better Chinese writing and voice preservation, not detector evasion.
Critical Rules
- Preserve facts. Do not add numbers, dates, sources, institutions, citations, people, claims, or quotes unless the user supplied them.
- If concrete evidence is missing, write
[需补事实] or ask for the missing fact instead of inventing it.
- Do not over-humanize. If the text already has a clear human voice, clean only formatting and obvious friction.
- Scenario first. Academic, official, legal, medical, and business writing must stay appropriately formal.
- Scores are editing signals, not proof of authorship.
Workflow
- Classify the task:
audit, rewrite, voice-match, or fact-safe rewrite.
- Identify scenario:
business, academic, official, wechat, zhihu, xhs, creative, or general.
- Run the CLI audit when a text file is available:
PYTHONPATH="$CLAUDE_SKILL_DIR/src" python3 -m renwei_zh audit "$FILE" --scenario "$SCENARIO" --json
- Read the highest-severity evidence first: fact-risk items, red hits, clusters, and model fingerprints.
- Rewrite structurally: delete formulaic scaffolding, replace abstract verbs with concrete user-supplied actions, vary sentence rhythm, and keep domain-appropriate terminology.
- Re-audit the revised text when practical.
- Output the final text plus a short evidence report: what changed, facts preserved, residual risks.
Load Router
- If the text is business, academic, official, platform-specific, or creative, read
references/05-vertical-scenarios.md.
- If the audit shows Claude, GPT, DeepSeek, Qwen, Gemini, or Agent-style fingerprints, read
references/03-llm-fingerprints.md.
- If many syntactic shells appear, read
references/02-syntactic-shells.md.
- If the user asks why something is considered AI-like, read
references/01-pattern-library.md.
- If the user asks for metrics or thresholds, read
references/04-statistical-features.md.
- If the user only wants command-line usage, read
references/06-detection-tools.md.
Golden Template
When rewriting, follow this output contract:
【门检】
- 场景:
- 是否需要重写:
- 事实风险:
【改写后】
...
【证据报告】
- 删除/改写的高风险模式:
- 保留的事实:
- 未确认但需要用户补充的事实:
- 残留风险:
Boundaries
NOT this skill: English prose editing, plagiarism, impersonation, fake citations, SEO spam, detector bypass guarantees, or rewriting content to hide misconduct.
1---2name: renwei-zh3description: Audit and revise Chinese text to reduce formulaic AI-style writing while preserving facts and author voice. Use for 中文润色, 去机器味, 去 AI 味, 公众号/知乎/小红书/商业/学术稿改写, 模型化表达审计, 声线保留, fact-safe Chinese editing.4---56# renwei-zh78Use this skill when the user wants Chinese text reviewed, revised, or audited for formulaic model-like writing. The objective is better Chinese writing and voice preservation, not detector evasion.910## Critical Rules1112- Preserve facts. Do not add numbers, dates, sources, institutions, citations, people, claims, or quotes unless the user supplied them.13- If concrete evidence is missing, write `[需补事实]` or ask for the missing fact instead of inventing it.14- Do not over-humanize. If the text already has a clear human voice, clean only formatting and obvious friction.15- Scenario first. Academic, official, legal, medical, and business writing must stay appropriately formal.16- Scores are editing signals, not proof of authorship.1718## Workflow19201. Classify the task: `audit`, `rewrite`, `voice-match`, or `fact-safe rewrite`.212. Identify scenario: `business`, `academic`, `official`, `wechat`, `zhihu`, `xhs`, `creative`, or `general`.223. Run the CLI audit when a text file is available:2324```bash25PYTHONPATH="$CLAUDE_SKILL_DIR/src" python3 -m renwei_zh audit "$FILE" --scenario "$SCENARIO" --json26```27284. Read the highest-severity evidence first: fact-risk items, red hits, clusters, and model fingerprints.295. Rewrite structurally: delete formulaic scaffolding, replace abstract verbs with concrete user-supplied actions, vary sentence rhythm, and keep domain-appropriate terminology.306. Re-audit the revised text when practical.317. Output the final text plus a short evidence report: what changed, facts preserved, residual risks.3233## Load Router3435- If the text is business, academic, official, platform-specific, or creative, read `references/05-vertical-scenarios.md`.36- If the audit shows Claude, GPT, DeepSeek, Qwen, Gemini, or Agent-style fingerprints, read `references/03-llm-fingerprints.md`.37- If many syntactic shells appear, read `references/02-syntactic-shells.md`.38- If the user asks why something is considered AI-like, read `references/01-pattern-library.md`.39- If the user asks for metrics or thresholds, read `references/04-statistical-features.md`.40- If the user only wants command-line usage, read `references/06-detection-tools.md`.4142## Golden Template4344When rewriting, follow this output contract:4546```text47【门检】48- 场景:49- 是否需要重写:50- 事实风险:5152【改写后】53...5455【证据报告】56- 删除/改写的高风险模式:57- 保留的事实:58- 未确认但需要用户补充的事实:59- 残留风险:60```6162## Boundaries6364NOT this skill: English prose editing, plagiarism, impersonation, fake citations, SEO spam, detector bypass guarantees, or rewriting content to hide misconduct.
zaoqu-liu