Skills
skills/jezweb/claude-skills/elevenlabs-agents/Gen Agent Trust Hub

elevenlabs-agents

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references several official packages from the ElevenLabs NPM scope for integration and management.
  • Evidence: Mentions @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, and @elevenlabs/agents-cli for local development.
  • [COMMAND_EXECUTION]: Provides administrative shell scripts for automating agent development workflows.
  • Evidence: scripts/create-agent.sh, scripts/deploy-agent.sh, and scripts/test-agent.sh utilize the official ElevenLabs CLI for project initialization and deployment.
  • [PROMPT_INJECTION]: Documentation includes extensive templates and frameworks for system prompts designed to define agent behavior and security boundaries.
  • Evidence: assets/system-prompt-template.md and references/system-prompt-guide.md provide structured components (Personality, Guardrails, Goals) to prevent unwanted agent behavior.
  • [SAFE]: The skill incorporates security-first architecture for production deployments.
  • Evidence: The 'Signed URLs' pattern in SKILL.md ensures that sensitive API keys are never exposed in client-side code, using server-side authentication proxies instead.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 10:38 PM