phone-verification-waterfall

Pass

Audited by Gen Agent Trust Hub on Mar 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted lead data from HubSpot contacts without explicit sanitization or boundary isolation. \n
  • Ingestion points: Lead data including names, emails, and job titles are pulled from the HubSpot CRM (SKILL.md, Stage 1). \n
  • Boundary markers: The prompt logic does not define clear boundaries or 'ignore instructions' delimiters for the data retrieved from external contact records. \n
  • Capability inventory: The skill is capable of performing batch updates to the HubSpot CRM and executing enrichment requests to third-party APIs (SKILL.md, Stages 2-4). \n
  • Sanitization: No specific sanitization or escaping mechanisms are described to prevent potentially malicious strings within contact fields from influencing the agent's behavior during classification or sync operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 22, 2026, 06:50 PM