post-demo-automation

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via Clari call intelligence: clari_get_call_summary extracts demo conversation points/objections from a recording whose content is authored by the customer/demo attendees (non-operating user), and that extracted prose is then used to generate the email/note.