Security audit
Security checks for vulnerabilities and agentic risk
This skill transparently helps users search for, inspect, and optionally install ClawHub skills using the ClawHub CLI.
Before installing any recommended skill, inspect it first and install only skills you trust, because adding skills can change what your agent can do in future sessions.
The trigger phrases are broad enough to match ordinary user requests about finding things, which can cause this skill to activate outside its intended scope. Because the skill recommends and installs third-party skills, accidental invocation can steer users toward unreviewed code or unnecessary installations, increasing supply-chain and prompt-routing risk.
The activation description says the skill should respond to general requests like '有没有 skill 可以...' and '我需要一个能...的 skill' without defining stricter boundaries or disambiguation rules. In context, this is risky because the skill's workflow culminates in inspecting and installing external skills, so ambiguous routing can lead users into unintended package discovery and potential supply-chain exposure.
The package name and description explicitly label the skill as "CN" / Chinese-focused, which indicates a locale-specific constraint. In this file, there is no natural-language explanation that the locale limitation is optional or justified, so it may conflict with policy expectations around language or locale choice.
The script includes user-facing text only in Chinese in both the header comment and usage message. This imposes a specific language/locale on users without opt-in or explanation, which matches the policy-violation category for language constraints.
No suspicious patterns detected.
