{
  "id": "benchmark:appworld",
  "_kind": "Benchmark",
  "_file": "benchmarks/benchmarks/benchmarks-agentic.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "displayName": "AppWorld",
    "homepageUrl": "https://appworld.dev/",
    "kind": "tool-use",
    "targetsKind": "AgentVersion",
    "description": "Tool-use benchmark with a controllable ecosystem of 9 simulated\napps (email, calendar, shopping, etc.) and 750 cross-app tasks.\nTests long-horizon multi-app orchestration.\n"
  },
  "outgoingEdges": [
    {
      "from": "benchmark:appworld",
      "to": "skill-area:multi-app-orchestration",
      "kind": "covers",
      "attributes": {
        "attributes": {
          "coverage": "full",
          "weight": 0.7
        }
      }
    },
    {
      "from": "benchmark:appworld",
      "to": "skill-area:multi-turn-tool-use",
      "kind": "covers",
      "attributes": {
        "attributes": {
          "coverage": "partial",
          "weight": 0.3
        }
      }
    },
    {
      "from": "benchmark:appworld",
      "to": "domain:software-engineering",
      "kind": "applies_to",
      "attributes": {}
    }
  ],
  "incomingEdges": []
}