and-clarify

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md requires reading repository/target work and State Reason via the configured backend contract (step 1: “Read the target work and latest State Reason”), which can include outsider-authored free text (e.g., issue/PR body, wiki/discussion content) that the LLM then uses during the interview.