Refactor Safely

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

• Third-party content exposure detected (high risk: 0.70). High risk: the workflow explicitly calls refactor_tool with mode="suggest" for “community-driven refactoring suggestions,” which plausibly ingests outsider-authored free-form text (e.g., community/package content or scraped suggestions) into the agent’s LLM context at runtime.