中文去AI味

v1.0.0

将AI生成的机械化文本转换为自然、有人情味的人类写作风格

2· 2.4k·36 current·36 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The name and description (humanize Chinese AI text) match the SKILL.md content: concrete rewriting strategies, examples, and checklists. There are no unrelated binaries, env vars, or config paths requested.
Instruction Scope
The runtime instructions are purely stylistic rewriting guidelines and do not instruct reading system files, contacting external endpoints, or accessing credentials. Note: the guidance explicitly encourages adding personal anecdotes and subjective language (e.g., “我之前也遇到过类似情况...”); that is coherent with the stated goal but can lead to fabricated personal experiences or unverifiable claims if applied blindly—review outputs for factual accuracy and appropriateness in sensitive contexts.
Install Mechanism
No install spec and no code files—this is instruction-only, which minimizes disk/network installation risk.
Credentials
The skill declares no required environment variables, credentials, or config paths. Nothing requests access to unrelated services or secrets.
Persistence & Privilege
always is false and the skill is user-invocable. The default ability for the agent to call the skill autonomously remains, but this is normal and not combined with elevated privileges or broad credential access.
Assessment
This skill appears coherent and low-risk: it only contains writing-style rules and examples and asks for nothing from your system. Before installing or using it, try these precautions: (1) run it on non-sensitive sample text to confirm the style change meets your needs; (2) be aware it encourages adding personal anecdotes and subjective claims—verify any factual statements and avoid generating real-world personal details that could be false; (3) avoid using it to produce legal, medical, or other high-stakes content without human review; (4) remember the agent may call the skill autonomously by default—monitor outputs initially to ensure it’s used appropriately.

Like a lobster shell, security has layers — review code before you run it.

latestvk979mz1dcm1jq300km6wch9v7h819pm6

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments