ln-611-docs-structure-auditor

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly directs the worker to fetch missing reference files via WebFetch from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path} and lists those reference files as MANDATORY READs that the agent must load and use to drive audit decisions, so public GitHub-hosted (user-generated) content can directly influence tool behavior.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill states that if the local shared/ directory is missing it will fetch required reference files at runtime from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}, and those fetched documents are declared "MANDATORY READ" (core contracts, scoring, templates) which directly control the auditor's prompts/behavior, so this external raw.githubusercontent.com URL is a required runtime dependency that can change agent instructions.