Deal Intelligence Summary

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from multiple external CRM sources without specific security boundaries.
  • Ingestion points: Data is pulled from get_deal, get_deal_activities, and list_tasks actions, and queried from tables including deal_health_scores and relationship_health_scores.
  • Boundary markers: No delimiters or instructions to ignore embedded commands within CRM fields (like activity descriptions or contact notes) are present in the instructions.
  • Capability inventory: The skill utilizes CRM tool access to retrieve and synthesize sales data.
  • Sanitization: There are no instructions for validating or sanitizing retrieved text before it is summarized.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 11:37 AM