image-gen

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via --promptfiles/stdin: scripts/main.ts reads arbitrary files/STDIN into prompt and passes it to provider APIs (e.g., Google/OpenAI/DashScope) as user text, which the provider LLM consumes; those prompt contents are not authored by the operating user.