ln-513-regression-checker

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md explicitly instructs that if shared/ is missing it should "fetch files via WebFetch from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}" and also mandates reading those shared/reference documents, meaning the agent will fetch and interpret public GitHub raw content (untrusted third‑party user-generated content) which can change detection, commands, and parsing behavior.

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 runtime reference files via WebFetch from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}, and those fetched documents are mandated "MANDATORY READ" references that directly influence the agent's detection, command selection, and output-normalization instructions, so the external URL controls agent behavior at runtime.