Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (朋友圈文案生成器) matches the manifest, SKILL.md and main.js: all define copy types and build prompt templates. Nothing in the files requests unrelated capabilities (no cloud creds, no system paths).
Instruction Scope
SKILL.md and main.js only construct and return a prompt template for an LLM; they do not themselves call external APIs or read files. This is coherent but important: the skill produces prompts for the agent to use — it does not directly generate final copy by itself. Users expecting the skill to call an external copy-generation API should note that the agent/host LLM will perform the actual generation.
Install Mechanism
No install spec, no downloads, and no dependencies. The skill is instruction-only with a small local JS file — minimal install risk.
Credentials
No required environment variables, no credentials, and no config paths. The skill does not ask for secrets or elevated access; requested surface is proportional to purpose.
Persistence & Privilege
always is false and the skill does not request persistent or elevated privileges. It does not modify other skills or system config.
Assessment
This skill is internally consistent and low-risk: it only constructs LLM prompts for writing WeChat Moments copy and does not access credentials or external endpoints. Two practical notes before installing: 1) the skill returns a prompt template — the actual text will be produced by your agent's LLM, so test it to confirm the final output meets your content and safety expectations; 2) the owner is unknown, so avoid feeding any sensitive or private data into the 'topic' field. Also review outputs for manipulation or policy issues (marketing/‘反认知’ styles can be persuasive) before publishing publicly.Like a lobster shell, security has layers — review code before you run it.
latestvk979gz9cp5jhaagyyj7vcjrqxs8255xv
License
MIT-0
Free to use, modify, and redistribute. No attribution required.