close-plan-generator

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text is ingested via runtime calls to Clari (clari_get_call_summary sentiment/objections/next steps) and HubSpot deal/contact notes (hubspot_get_deal / hubspot_get_contact), which can include messages authored by non-operating-user participants (e.g., prospect/customer communications) that are then placed into the agent’s LLM context for close-plan generation.