use-local-whisper
Warn
Audited by Snyk on Jun 19, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.75). The skill’s runtime path is local on-device transcription (downloading WhatsApp audio via Baileys and running
whisper-cli), but the LLM context injection risk depends on whether the resulting transcript (outsider-authored WhatsApp voice note text) is fed into the agent’s LLM; the SKILL.md indicates it will produce[Voice: <transcript>], which is free text authored by non-operating users.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata