ln-635-test-isolation-auditor

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

• Third-party content exposure detected (high risk: 1.00). This skill's SKILL.md says that if local shared/ files are missing it will fetch mandatory "shared/references/..." files via WebFetch from the public GitHub raw URL (https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}), and those mandatory reads are parsed and used to drive audit rules, scoring, and decisions, so untrusted third-party content can materially influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs at runtime to WebFetch mandatory reference files from https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/master/skills/{path}, and those fetched "MANDATORY READ" documents are required and can directly control the agent's instructions/output format.