skills/blockrunai/franklin/phone-call/Gen Agent Trust Hub

phone-call

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a robust workflow for placing AI-driven phone calls using platform-provided tools.
  • [SAFE]: Human-in-the-loop confirmation is required before the VoiceCall tool is invoked. The agent must show the user a summary of the call details and wait for an explicit 'yes' confirmation, which is a strong mitigation against automated abuse.
  • [SAFE]: The instructions include specific compliance guidelines (TCPA, daytime calling restrictions) and strictly limit the operational scope to US and Canadian phone numbers (+1 country code).
  • [SAFE]: The skill explicitly forbids anti-patterns such as calling emergency services, using manipulative scripts, or performing sequential cold calls without individual consent.
  • [SAFE]: External communication is handled entirely through the platform's native tools (ListPhoneNumbers, BuyPhoneNumber, VoiceCall, VoiceStatus) for interacting with the service provider.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 08:17 PM
Security Audit — agent-trust-hub — phone-call