business-kickoff-workshop

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

• Third-party content exposure detected (high risk: 0.85). The skill’s Phase 1 runtime path ingests outsider-authored free text when the operating user pastes “Fireflies transcripts, sales call notes, rough bullets” from the customer into the agent (“Share whatever context you have…”), and that pasted transcript/notes are then used to generate the customized agenda in the LLM context.