sfd-media-critic

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md defines a POST /v1/watch that accepts arbitrary "mediaUrl" inputs and instructs the agent to analyze/verify content and take actions (quality checks, regenerate, change prompts/models), meaning untrusted public/user-generated media could be ingested and its content could materially influence the agent's decisions and tool use.