fal-generate

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

• Third-party content exposure detected (high risk: 0.80). The skill accepts arbitrary public URLs as inputs (see SKILL.md "Option 3: Use existing URL — Any public URL works" and scripts/generate.sh's --image-url) which are passed to the generation service and thus cause the agent/backend to fetch and interpret untrusted third‑party content.