iiRecord
Agentic AI Atlas · AndroidWorld programmatic task suite
test-set:androidworld-programmatic-tasksa5c.ai
II.
TestSet JSON

test-set:androidworld-programmatic-tasks

Structured · live

AndroidWorld programmatic task suite json

Inspect the normalized record payload exactly as the atlas UI reads it.

File · benchmarks/test-sets/source-backed-benchmark-test-sets.yamlCluster · benchmarks
Record JSON
{
  "id": "test-set:androidworld-programmatic-tasks",
  "_kind": "TestSet",
  "_file": "benchmarks/test-sets/source-backed-benchmark-test-sets.yaml",
  "_cluster": "benchmarks",
  "attributes": {
    "displayName": "AndroidWorld programmatic task suite",
    "benchmarkId": "benchmark:android-world",
    "caseCount": 116,
    "composition": "AndroidWorld's reproducible mobile-agent task suite covers real\nAndroid apps with programmatic rewards and dynamically-instantiated\nnatural-language tasks.\n",
    "homepageUrl": "https://google-research.github.io/android_world/",
    "description": "Canonical AndroidWorld artifact for autonomous Android UI-control\nevaluation.\n"
  },
  "outgoingEdges": [
    {
      "from": "test-set:androidworld-programmatic-tasks",
      "to": "benchmark:android-world",
      "kind": "belongs_to_benchmark"
    }
  ],
  "incomingEdges": []
}