find-skills

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). This skill’s required workflow instructs the agent to run the Skills CLI (npx skills find '[query]') which, at runtime, queries the open skills ecosystem and returns skill metadata from third-party packages (outsider-authored free text) that the agent then presents/uses in its LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The README includes an explicit runtime installation command that clones remote code which will be installed as a skill and thus can control agent behavior: https://github.com/KimYx0207/Kim_Service.git