prompt-hub
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill incorporates proactive security measures as 'load-bearing invariants,' specifically requiring a 'Red Team' status check and an automated 'Voice gate' (lib/voice/frankx-voice.ts) to audit outputs for banned content.
- [SAFE]: External references are limited to well-known technology providers and official documentation from trusted organizations such as Anthropic, Google, and OpenAI, as well as established open-source tools like Promptfoo and Fabric.
- [SAFE]: The skill uses a structured routing mechanism via a conductor agent, reducing the risk of unauthorized tool use or accidental exposure by enforcing canonical workflows.
- [SAFE]: No instances of prompt injection, obfuscation, or data exfiltration attempts were found during the analysis of the skill's logic and instructions.
Audit Metadata