speech-to-text
Fail
Audited by Snyk on Nov 13, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (medium risk: 0.30). The skill is a legitimate ElevenLabs transcription guide, but its setup line tells the agent to run an unpinned
npx -y @elevenlabs/cli@latest onboard initthat signs the user in and stores a key, a broad action the user may not expect, so under the recall-first rule it is rated medium rather than no.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required workflow transcribes audio or video that the user or agent selects (local files,
source_url,cloud_storage_url, or a supplied stream URL), so outsider speech reaches the LLM only after an explicit selection of arbitrary media, which is the low-risk case.
Issues (2)
E004
CRITICALPrompt injection detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata