codex-auto-clip-workflow

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text via an externally provided reference video/“script.txt” used to generate voiceover/subtitles (Prompts 1 and 3), meaning a user can supply poison text to be read by the LLM during VO script handling.