multi-ai-agent-security-sandbox

Execute untrusted code or commands in a secure, isolated Docker environment. Use when asked to run unknown scripts, third-party skills, or any code that might pose a security risk to the host system.