hypeauditor

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to call HypeAuditor via Membrane (e.g., "membrane action run ..." and "membrane request CONNECTION_ID /path/to/endpoint"), which fetches data from a public third‑party API (public influencer/social data) that the agent is expected to read and use to drive actions and decisions, exposing it to untrusted, user‑sourced content.