call-prep

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

• Third-party content exposure detected (high risk: 0.90). The skill's Execution Flow (Step 2: Research Supplement) explicitly instructs the agent to "always run (web search)" and fetch public sources such as "[Company] news", funding announcements, leadership changes and attendee LinkedIn profiles—untrusted, user-generated or public web content that the agent is expected to read and synthesize into agenda/decisions, creating a clear vector for indirect prompt injection.