agents
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill guides users to install official ElevenLabs packages and tools from established registries (NPM and PyPI), including
@elevenlabs/cli,@elevenlabs/elevenlabs-js, andelevenlabs. - [REMOTE_CODE_EXECUTION]: Provides instructions for embedding a voice AI widget using a remote script from a well-known CDN (
https://unpkg.com/@elevenlabs/convai-widget-embed). This is a standard service integration pattern for the platform. - [INDIRECT_PROMPT_INJECTION]: The voice agents configured via this skill are designed to process user audio and text input. The documentation provides specific guidance on implementing security guardrails, including prompt injection detection and content moderation, to protect the agent context.
- [CREDENTIALS_UNSAFE]: The skill references sensitive configuration items such as API keys and local credential storage paths (
~/.agents/api_keys.json). These references are strictly documentation of standard security practices and use placeholders rather than exposing actual secrets.
Audit Metadata