Not in the top 1,000 by GitHub stars
baibanbao/qu-ai-wei
去 AI 味：中文写作的 AI 痕迹清理 skill。合并三家规则（含语料实测），并裁决它们互相矛盾的七处。A Claude skill for stripping AI tells from Chinese drafts.
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
SKILL.md
Week of Nov 2, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → dfc0d6a2627f
No earlier version was observed for comparison.
73GitHub stars
0stars in 7 days
1skill in this repository
0commits in 30 days
Heat
0.3- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
30-day change not yet available: not enough history.
Individual skills in this repository
How it works
qu-ai-wei
When your agent uses it
王佩专用的去 AI 味系统。触发词：去AI味、去味。合并三家——王佩写作系统 v4.0、humanizer-zh、lieflat-less-ai-tone（含实测数据）。先选模式（自家稿 / 白名单），再过硬边界，再下刀。不判断作者身份，不保证通过 AI 检测器。
What it loads into context
Frontmatter
94 tokensAlways available to the agent
SKILL.md
7,526 tokensWhen the skill is activated
Reference documents
1,779 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 三家分工
- 〇 先选模式
- 一 硬边界（优先于所有去味规则）
- 二 病根三条
- 三 门槛（机械层，先于一切）
- 四 有实测依据的十一刀
- 五 老编辑的刀
- 六 黑名单
- 七 不许动
- 八 三家打架时怎么裁
- 九 加回来（只在自家稿模式）
- 十 分文体的加戏位置
- 十一 声音校准（只在自家稿模式）
- 十二 报告格式（自家稿模式默认）
- 十三 交付前核对
- 边界
- 来源与致谢
Files
.gitignoreasset16 bytesLICENSEasset1,075 bytesREADME.mddoc6,302 bytesSKILL.mdcore24,979 bytes
Read SKILL.md
# 去 AI 味 v1.0
## 三家分工
| 来源 | 管什么 | 本 skill 取了什么 |
|------|--------|------------------|
| 王佩写作系统 v4.0 | 像不像人写的 | 病根三条、语言禁区、声音校准、B 级档位 |
| humanizer-zh | 删的时候别删坏 | 信息守恒、保留边界、交付前核对 |
| lieflat-less-ai-tone | 哪些是真痕迹 | 十一条实测规则、"不作为改写理由"表 |
王佩的系统和 humanizer-zh 靠经验，lieflat 靠语料（300 篇 AI 样本、117.9 万汉字、5 个模型）。三家打架时看 §八。
---
Skills included
- qu-ai-wei
Repository facts
- License
- MIT
- Created
- Sep 23, 2026
- Last pushed
- Sep 23, 2026
- Forks
- 15
- Open issues
- 0
- Project activity
- Active
Before you install
Instructions onlyAutomated signals, not a guarantee
Checked Nov 5, 2026
- No broad tool permissions
- No dangerous shell commands
- 9 network referencesin 2 files: README.md:16 SKILL.md:410
- No hidden or obfuscated text
- No instruction overrides
- No executable scripts
- No access to secrets
- No requests to vote in Hot
Show all 9 signals by file
README.md
- Network accessline 16
|------|--------|----------| | 王佩写作系统 v4.0 | 像不像人写的 | 病根三条、语言禁区、声音校准、协作档位 | | [humanizer-zh](https://github.com/op7418/Humanizer-zh) | 删的时候别删坏 | 信息守恒、保留边界、交付前核对 | - Network accessline 17
| 王佩写作系统 v4.0 | 像不像人写的 | 病根三条、语言禁区、声音校准、协作档位 | | [humanizer-zh](https://github.com/op7418/Humanizer-zh) | 删的时候别删坏 | 信息守恒、保留边界、交付前核对 | | [lieflat-less-ai-tone](https://github.com/larashero3-dotcom/lieflat-less-ai-tone) | 哪些是真痕迹 | 十一条实测规则、「不作 - Network accessline 18
| [humanizer-zh](https://github.com/op7418/Humanizer-zh) | 删的时候别删坏 | 信息守恒、保留边界、交付前核对 | - Network accessline 19
| [lieflat-less-ai-tone](https://github.com/larashero3-dotcom/lieflat-less-ai-tone) | 哪些是真痕迹 | 十一条实测规则、「不作为改写理由」表 | - Network accessline 60
git clone https://github.com/baibanbao/qu-ai-wei.git ~/.claude/skills/qu-ai-wei - Network accessline 89
It merges three rule sets — the author's own writing system, plus the open-source [humanizer-zh](https://github.com/op7418/Humanizer-zh) and [lieflat-less-ai-tone](https://github.com/larashero3-dotcom/lieflat-less-ai-tone) — and adjudicates
SKILL.md
- Network accessline 410
- **王佩写作系统 v4.0**——作者自用 skill，未公开。本 skill 取其病根三条、语言禁区、声音校准、协作档位。 - **humanizer-zh**——<https://github.com/op7418/Humanizer-zh>（MIT）。取其信息守恒约束、保留边界、交付前核对。该项目本身基于 blader/humanizer v3.0.0 与其 PR 39。 - Network accessline 411
- **humanizer-zh**——<https://github.com/op7418/Humanizer-zh>（MIT）。取其信息守恒约束、保留边界、交付前核对。该项目本身基于 blader/humanizer v3.0.0 与其 PR 39。 - Network accessline 412
- **lieflat-less-ai-tone**——<https://github.com/larashero3-dotcom/lieflat-less-ai-tone>（MIT）。取其十一条实测规则与「不作为改写理由」表。文中所有频率数据（每千字、倍率、覆盖率）均出自该项目的语料实测，方法与脚本见其 RESEARCH.md。
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Signal confidence: 9 facts
- Network access in prose of documentationREADME.md:16Low confidence
- Network access in prose of documentationREADME.md:17Low confidence
- Network access in prose of documentationREADME.md:18Low confidence
- Network access in prose of documentationREADME.md:19Low confidence
- Network access in code of documentationREADME.md:60Low confidence
- Network access in prose of documentationREADME.md:89Low confidence
- Network access in prose of the SKILL.mdSKILL.md:410Low confidence
- Network access in prose of the SKILL.mdSKILL.md:411Low confidence
- Network access in prose of the SKILL.mdSKILL.md:412Low confidence
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Compatible | Standard SKILL.md SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
