voice-ux-pro
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Prompt Injection (LOW): The prompt template in references/1-low-latency-voice-stack.md interpolates untrusted user input directly without delimiters. 1. Ingestion points: The [TEXT] variable in the LLM prompt pattern receives user voice data. 2. Boundary markers: Absent. No quotes, triple backticks, or special delimiters are used to isolate user input from the system instructions. 3. Capability inventory: The skill manages audio playback, emotive synthesis, and haptic feedback. 4. Sanitization: No sanitization or escaping logic is described for the user-provided content.
Audit Metadata