ln-913-community-debater

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

• Third-party content exposure detected (high risk: 0.90). The skill's Phase 0 "MANDATORY READ" steps and the top-level fallback explicitly instruct the agent to fetch public files via WebFetch from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path} and to load target-repository docs, meaning it ingests untrusted, user-generated GitHub content that can materially influence decisions (repo IDs, categories, discussion text, and publishing actions).

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly fetches required "MANDATORY READ" files at runtime from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}, and those fetched documents (formatting, discovery, humanizer checklist) directly control the agent's prompt/instruction generation, so remote content can alter or control behavior.