AGENT LAB: SKILLS

fal-audio

Warn

Audited by Snyk on Feb 15, 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). The skill's speech-to-text flow (scripts/speech-to-text.sh and the MCP example using audio_url) accepts arbitrary public audio URLs and transcribes that untrusted, user-provided content via the fal.run API, which the agent reads and returns—exposing it to indirect prompt injection.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 08:27 PM