create-issue

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

• Third-party content exposure detected (high risk: 0.75). In --interview mode the workflow reads arbitrary repository files at runtime (git ls-files + README + entry-point files) and feeds their contents into the LLM context for targeted questions, which can include outsider-authored free text (e.g., codebase/README written by others).