ln-740-quality-setup

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches missing shared files from the public URL pattern https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/{path} and even requires loading shared/references/meta_analysis_protocol.md, meaning it ingests untrusted, user-hosted GitHub content that could influence tool invocation and decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly performs a runtime WebFetch of required shared files from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/{path}, and those fetched reference files (e.g., meta_analysis_protocol.md) are required and read as instructions that directly influence agent behavior.