image-workflow

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

• Third-party content exposure detected (high risk: 0.80). The workflow explicitly calls the sub-skill "rebyteai/internet-search" in Step 2 ("Research References (if needed)"), instructing the agent to fetch and use public web references (brand logos, artists, real products, trends), which are untrusted third-party sources that the agent will read and act on to inform generation decisions.