ship

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

• Third-party content exposure detected (high risk: 0.72). Outsider free text can enter the LLM context via Phase 4 “Bot Watch” and Phase 5 “Address Bot Feedback,” where the skill performs GraphQL queries to fetch PR review/comments authored by other users (e.g., bot or human reviewers) and then uses that content for the LLM-driven addressing loop.