visual-ops-writer

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow fetches external article URLs via fetch_article.py without the agent first selecting a specific item, allowing outsider-submitted text from arbitrary URLs.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The script scripts/generate_image.py uses https://redfox.hk/story/api/parseWork/imageGen/submitSkill and other endpoints at redfox.hk at runtime to submit, upload, and poll image generation tasks, but redfox.hk is an arbitrary/uncontrolled third-party platform lacking verified official enterprise provenance.