logo-and-wordmark-design

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

• Third-party content exposure detected (high risk: 0.70). The required workflow is driven by “Brand owner or brief / product/marketing teams” inputs (e.g., brand strategy, name, USP, usage contexts), which are outsider-authored free-form text that the agent would need to read and incorporate into the LLM context to proceed.