walkie-talkie

Warn

Audited by Snyk on Mar 1, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.80). This skill's SKILL.md explicitly ingests WhatsApp voice-note media (e.g., "MEDIA: /path/to/file") and uses local Whisper transcription treating the resulting text as the user's input, so untrusted user-generated audio could contain instructions that influence agent behavior.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 1, 2026, 01:38 AM