account-handover

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the query-run MCP tool to execute SQL queries on internal PostgreSQL databases (postgres.prod.billing_usagereport) to retrieve usage metrics and billing trends.
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by ingesting unstructured data from customer communications.
  • Ingestion points: The skill reads external account notes and conversation logs via the vitally:get_account_notes and vitally:get_account_conversations tools.
  • Boundary markers: There are no explicit delimiters or instructions within the skill logic to prevent the agent from following commands embedded within the retrieved notes.
  • Capability inventory: The skill has access to internal usage databases through SQL execution and the Vitally CRM API.
  • Sanitization: No sanitization, escaping, or validation logic is present to process the external text before it is included in the final handover document.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 11:45 AM