Humanizer Zh
v1.0.0去除文本中的 AI 生成痕迹。适用于编辑或审阅文本，使其听起来更自然、更像人类书写。 基于维基百科的"AI 写作特征"综合指南。检测并修复以下模式：夸大的象征意义、 宣传性语言、以 -ing 结尾的肤浅分析、模糊的归因、破折号过度使用、三段式法则、 AI 词汇、否定式排比、过多的连接性短语。
⭐ 0· 123·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (去除 AI 写作痕迹) matches the SKILL.md and README. The skill requires no binaries, no env vars, and its allowed tools (Read/Write/Edit/AskUserQuestion) are appropriate for a text-editing/rewrite helper.
Instruction Scope
SKILL.md contains detailed, prescriptive editing rules and examples that stay within the stated purpose. It does instruct the agent to read and rewrite supplied text and includes examples that reference processing files (e.g., article.md). This is expected, but users should be aware the skill is designed to read file contents when invoked.
Install Mechanism
The skill bundle itself has no formal install spec (instruction-only), which is low risk. The README suggests installation via npx or git clone from a GitHub repo (op7418). Fetching code from an external GitHub repo is normal but worth verifying the repository/owner before running installation commands.
Credentials
No environment variables, credentials, or config paths are requested. The lack of secrets or unrelated credentials is proportionate to a text-editing skill.
Persistence & Privilege
Flags show always: false and default autonomous invocation. The skill does not request persistent system presence or modify other skills or system-wide settings in the provided files.
Assessment
This skill appears to be what it says: a rule-driven Chinese 'humanizer' that rewrites text to remove AI writing patterns. Before installing, verify the GitHub source/owner referenced in the README (the package suggests cloning npx/git clone from op7418) so you trust the code you download. Remember that when invoked it can read and edit text files you point it at (e.g., article.md), so avoid passing sensitive files. Autonomous invocation is allowed by default on the platform (normal), but you can restrict usage or inspect the SKILL.md and README locally first if you prefer. If you need stronger assurance, review the remote repository contents or ask the author for provenance before adding the skill.Like a lobster shell, security has layers — review code before you run it.
latestvk970ynt5z0eczb51yg3f4w4n0h84gm7d
License
MIT-0
Free to use, modify, and redistribute. No attribution required.