de-ai-writerSAFE
Rewrite AI-generated copy into natural human tone. AI文案去AI味,自然化改写。
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Rewrite AI-sounding Chinese copy into text that reads like a human wrote it.
把 AI 生成的中文改写成有真人味的文字:专治空洞拔高、排比三连、官方黑话、模板化结尾。适用于公众号、小红书、电商文案、产品描述。
[](LICENSE) [](#install) [](https://github.com/jiawood2006/hermes-skills)
Install
# 1) Agent Skills CLI (works with 50+ agent hosts) npx skills add jiawood2006/de-ai-writer # 2) Claude Code / any host that reads ~/.claude/skills git clone https://github.com/jiawood2006/de-ai-writer.git ~/.claude/skills/de-ai-writer # 3) Hermes Agent git clone https://github.com/jiawood2006/de-ai-writer.git ~/.hermes/skills/utilities/de-ai-writer
No build step, no account, no telemetry.
Usage
# A) with an API key (best quality) — any OpenAI-compatible endpoint export LLM_API_KEY="sk-..." # or pass --api-key export LLM_BASE_URL="https://api.deepseek.com/v1" export LLM_MODEL="deepseek-chat" python3 scripts/deai.py 稿子.txt -o 改好.txt # rewrite a file python3 scripts/deai.py 稿子.txt # print to stdout # B) zero-cost mode — no API call at all, just prints a rewrite prompt you can paste anywhere python3 scripts/deai.py 稿子.txt --prompt-only
Ask your agent: "帮我把这段去一下 AI 味" and it will drive the script for you.
Requirements
Python 3.9+. Zero third-party dependencies (standard library only). LLM_API_KEY is optional — without it use --prompt-only.
Notes
The zero-cost mode exists on purpose: you can evaluate the rewriting logic before spending a single token.
License
MIT — see LICENSE. Issues and PRs welcome. Part of the hermes-skills collection.
f2a039bb8e11OBSERVED · 2026-10-13Install
Commands as the repository documents them. They are shown, not run.
npx skills add jiawood2006/de-ai-writer
git clone https://github.com/jiawood2006/de-ai-writer.git ~/.claude/skills/de-ai-writer
git clone https://github.com/jiawood2006/de-ai-writer.git ~/.hermes/skills/utilities/de-ai-writer
Host compatibility
What the documentation claims. We have not run a compatibility test.
| Host | Status | Notes |
|---|---|---|
| claude-code | mentioned |
What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
---
name: de-ai-writer
description: "AI 文案去味器。用户提供 AI 生成的文本(公众号/小红书/电商文案等),觉得"太 AI 味"要改得更自然、更像人写的时使用。输出自然有人味的中文。De-AI writer: rewrite AI-generated copy (WeChat articles, Xiaohongshu, e-commerce) into natural, human-sounding Chinese text."
version: 1.0.0
author: 涛哥
license: MIT
metadata:
hermes:
tags: [writing, editing, humanize, de-ai, copywriting]
category: utilities
homepage: https://gitee.com/tao6677/useful-tools
---
# De-AI Writer AI 文案去味器
把 AI 生成的文字改写成自然、有真人味的中文。专治空洞拔高、排比三连、官方黑话、模板化结尾。
## 触发条件
用户提供一段文字(粘贴或文件),要求:
- "去AI味""改得像人写的""太官方了改自然点"
- 润色文案(公众号/小红书/电商标题/产品描述)
## 使用步骤
### 1. 有 API Key(推荐,效果好)
```bash
export LLM_API_KEY="你的key"
export LLM_BASE_URL="https://api.deepseek.com/v1" # 任意 OpenAI 兼容接口
export LLM_MODEL="deepseek-chat"
python3 ~/.hermes/skills/utilities/de-ai-writer/scripts/deai.py 稿子.txt -o 改好.txt
```
### 2. 无 API Key(零成本模式)
```bash
python3 ~/.hermes/skills/utilities/de-ai-writer/scripts/deai.py 稿子.txt --prompt-only
```
输出内置完整改写规则的提示词,粘贴到任何 AI 助手(ChatGPT/文心/Kimi)即可改写。
## 零依赖
纯 Python 标准库,无需安装任何包。Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (3 observation(s))
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
f2a039bb8e11full audit observations/trust-audit/skill/jiawood2006__de-ai-writer.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-13 | f2a039bb8e11 | SAFE | B | 89 | first audit |
Questions
What does the de-ai-writer skill do?
Rewrite AI-generated copy into natural human tone. AI文案去AI味,自然化改写。
Is de-ai-writer safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can de-ai-writer access on my machine?
The audit observed that it reaches the network. Each of those is consistent with what it says it does. Secrets in the source: none found.
Which assistants does de-ai-writer work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (f2a039bb8e11), read on 2026-10-13. The repository is watched, and a new audit runs when it changes — this is the first audit.