elevenlabs-agents
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and installs official ElevenLabs packages such as @elevenlabs/client, @elevenlabs/react, and @elevenlabs/agents-cli. These are well-known services from a technology provider, and their usage is standard for the skill's purpose.
- [COMMAND_EXECUTION]: Shell scripts are included to automate agent creation and testing using the official ElevenLabs CLI. These scripts are benign utilities that verify authentication before execution and do not contain suspicious command injections.
- [CREDENTIALS_UNSAFE]: The documentation and boilerplate code correctly utilize environment variables (e.g., ELEVENLABS_API_KEY) and platform-specific secret placeholders (e.g., {{secret__xxx}}) to handle credentials securely. No hardcoded secrets were identified.
- [PROMPT_INJECTION]: The skill provides structured guidance on prompt engineering, including the use of guardrails and character constraints to keep agents within safety boundaries. No instructions were found that attempt to bypass AI safety filters.
Audit Metadata