afk-cook

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

• Third-party content exposure detected (high risk: 0.90). Yes — the required prompt template (slice-prompt.md) explicitly tells the agent to pull and read GitHub issue bodies/parent PRDs via gh issue view, which are user-generated, public third-party contents the agent will interpret and act on (influencing commits, tests, and next actions), enabling indirect prompt injection.