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": "ART",
  "description": "Train LLMs to be better agents using RL",
  "url": "https://art.openpipe.ai/",
  "version": "1.0.0",
  "protocolVersion": "0.3",
  "preferredTransport": "HTTP+JSON",
  "supportedInterfaces": [
    {
      "url": "https://art.openpipe.ai/",
      "protocolBinding": "HTTP+JSON",
      "protocolVersion": "0.3"
    }
  ],
  "provider": {
    "url": "https://art.openpipe.ai/",
    "organization": "ART"
  },
  "documentationUrl": "https://art.openpipe.ai/",
  "capabilities": {
    "streaming": false,
    "pushNotifications": false
  },
  "defaultInputModes": [
    "text/plain"
  ],
  "defaultOutputModes": [
    "text/plain"
  ],
  "skills": [
    {
      "id": "openpipe",
      "name": "Openpipe",
      "description": "Use when training LLM-based agents to improve performance through reinforcement learning. Reach for this skill when building agents that need to learn from experience, optimize tool usage, handle multi-step reasoning, or fix specific behavioral issues. Apply ART when you have a task that agents can attempt repeatedly, can be scored objectively or with an LLM judge, and where smaller models should outperform larger ones.",
      "tags": [],
      "url": "https://art.openpipe.ai/.well-known/agent-skills/openpipe/skill.md"
    }
  ]
}