Run coding agents in isolated observable VibeKit sandboxes
Launch coding agents such as Claude Code, Gemini, or Codex in clean sandboxed sessions with redaction and run observability.
npx skills add agentskillexchange/skills --skill run-coding-agents-in-isolated-observable-vibekit-sandboxes
Use this skill when an operator wants an AI coding agent to work inside a controlled session rather than directly on a sensitive host. The workflow starts a VibeKit sandbox, configures the chosen coding agent, applies redaction and observability controls, lets the agent run, and reviews the resulting trace and code changes.
What this skill actually does
Invoke it instead of running the coding agent normally when isolation, secrets protection, or auditable execution traces matter more than raw convenience.
Scope boundary: this is not a generic coding-agent framework listing. It is specifically about supervised sandbox execution, redaction, and observability for existing coding agents.