ln-162-skill-reviewer

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs that if shared/ is missing the agent should "fetch files via WebFetch from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}", which causes the agent to ingest public GitHub content (untrusted third-party, user-provided) that the workflow then MANDATORY READs and uses to drive checks and decisions.

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

• Potentially malicious external URL detected (high risk: 0.80). The SKILL explicitly instructs fetching missing shared files at runtime from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}, which is a remote raw-GitHub URL used to load required MANDATORY READ content that can directly control agent prompts/instructions, so this is a runtime external dependency that poses risk.