agents
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents the use of the ElevenLabs CLI for project management, agent configuration, and procedure deployment. These commands are standard for the platform's development workflow.
- [EXTERNAL_DOWNLOADS]: References official ElevenLabs packages for Node.js and Python registries, as well as a widget embed script from a well-known CDN (unpkg). All resources are official vendor tools.
- [REMOTE_CODE_EXECUTION]: Provides a shell installer script for the ElevenLabs CLI hosted on the vendor's official GitHub repository. Per security rules, this is considered safe as it originates from the verified vendor.
- [PROMPT_INJECTION]: Includes extensive documentation on platform-level guardrails (
platform_settings.guardrails) designed to detect and block prompt injection and instruction override attempts independently of the LLM. - [CREDENTIALS_UNSAFE]: Recommends secure handling of API keys through environment variables and an onboarding process that utilizes the OS keyring for credential storage, avoiding hardcoded secrets.
Audit Metadata