elevenlabs-calls
Warn
Audited by Socket on Jul 15, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: The skill is purpose-aligned and uses official service endpoints, with no notable installer or proxy abuse. However, it grants an AI agent the ability to initiate real-world outbound calls and relies on raw local credential files, which is high-impact and insufficiently scoped without strong approval and consent controls.
Confidence: 87%Severity: 74%
Audit Metadata