Deal Handoff Brief

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE]: The skill intentionally accesses sensitive information from a CRM, including deal values, stakeholder contact details, meeting transcripts, and internal strategy notes. This data is used exclusively to populate the handoff brief and transition email as requested by the user.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources (CRM activity logs, meeting notes, and stakeholder roles) and interpolates this content into the generated handoff brief.
  • Ingestion points: Data is ingested via execute_action calls in SKILL.md (e.g., get_deal_activities, get_meetings).
  • Boundary markers: The skill does not explicitly define delimiters to isolate untrusted CRM content from the agent's instructions.
  • Capability inventory: The skill utilizes the crm capability to read and process records.
  • Sanitization: There is no evidence of sanitization or filtering for the data retrieved from the CRM fields.
  • [SAFE]: All external resources and tool calls (e.g., execute_action) are consistent with the vendor's (SixtySecondsApp) infrastructure and intended sales-enablement use case.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 11:37 AM