startup-channel-finder

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

• Third-party content exposure detected (high risk: 0.85). The required workflow instructs the agent to perform “current web research” (including opening “original pages”) and then to place the resulting extracted text/evidence into the analysis JSON, which is later rendered into the LLM/agent-visible report; this can include outsider-authored free text from fetched public web pages.