Card snapshot

This is a frozen copy of the agent's agent-card.json as we observed it at the timestamp above. We capture a new snapshot every time the card's content hash changes. Useful for: forensic drift analysis, verifying downstream callers see the right version, reproducing routing decisions made historically.

{
  "name": "microsandbox",
  "description": "Easy, fast, local microVMs for untrusted workloads",
  "url": "https://docs.microsandbox.dev/",
  "version": "1.0.0",
  "protocolVersion": "0.3",
  "preferredTransport": "HTTP+JSON",
  "supportedInterfaces": [
    {
      "url": "https://docs.microsandbox.dev/",
      "protocolBinding": "HTTP+JSON",
      "protocolVersion": "0.3"
    }
  ],
  "provider": {
    "url": "https://docs.microsandbox.dev/",
    "organization": "microsandbox"
  },
  "documentationUrl": "https://docs.microsandbox.dev/",
  "capabilities": {
    "streaming": false,
    "pushNotifications": false
  },
  "defaultInputModes": [
    "text/plain"
  ],
  "defaultOutputModes": [
    "text/plain"
  ],
  "skills": [
    {
      "id": "microsandbox",
      "name": "Microsandbox",
      "description": "Use when running untrusted code in isolated microVMs \u2014 AI agents, user submissions, plugins, CI jobs, package installs, and generated code. Use for creating sandboxes programmatically via SDK (Rust, TypeScript, Python, Go) or managing them from the CLI. Use when you need hardware-isolated execution with controlled network access, secret injection, and filesystem isolation.",
      "tags": [],
      "url": "https://docs.microsandbox.dev/.well-known/agent-skills/microsandbox/skill.md"
    }
  ]
}