{
  "id": "test-set:appworld-750-cross-app-tasks",
  "_kind": "TestSet",
  "_file": "benchmarks/test-sets/source-backed-benchmark-test-sets.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "displayName": "AppWorld 750 cross-app tasks",
    "benchmarkId": "benchmark:appworld",
    "caseCount": 750,
    "composition": "AppWorld's benchmark task suite exercises natural, diverse, and\nchallenging autonomous-agent tasks across a simulated world of\nday-to-day apps, APIs, and user data.\n",
    "homepageUrl": "https://appworld.dev/",
    "description": "Canonical AppWorld artifact for rich interactive code-generation\nand multi-app tool-use evaluation.\n"
  },
  "outgoingEdges": [
    {
      "from": "test-set:appworld-750-cross-app-tasks",
      "to": "benchmark:appworld",
      "kind": "belongs_to_benchmark"
    }
  ],
  "incomingEdges": []
}