{
  "aliases": [],
  "authors": [],
  "details": {
    "benchmark_protocol": {
      "actions": [
        "harness-defined browser actions through the task's local browser service"
      ],
      "counts": [
        {
          "basis": "reported",
          "evidence_url": "https://raw.githubusercontent.com/TIGER-AI-Lab/WebsiteBench/78f5558e8197bb91b3bab63428c12b92060ddc6f/docs/tasks.md",
          "population": "released tasks",
          "qualifier": "exact",
          "unit": "tasks",
          "value": 70
        },
        {
          "basis": "computed",
          "evidence_url": "https://raw.githubusercontent.com/TIGER-AI-Lab/WebsiteBench/78f5558e8197bb91b3bab63428c12b92060ddc6f/docs/tasks.md",
          "population": "hidden test cases summed over the docs/tasks.md table",
          "qualifier": "exact",
          "unit": "test cases",
          "value": 46916
        }
      ],
      "evaluation": "Hidden test cases per task. A case passes when its functional checks pass and every visual checkpoint's flat-baseline-normalized RGB SSIM reaches 0.99. A task's reward is passed cases divided by cases, weighted equally.",
      "evidence_urls": [
        "https://raw.githubusercontent.com/TIGER-AI-Lab/WebsiteBench/78f5558e8197bb91b3bab63428c12b92060ddc6f/README.md",
        "https://raw.githubusercontent.com/TIGER-AI-Lab/WebsiteBench/78f5558e8197bb91b3bab63428c12b92060ddc6f/docs/tasks.md"
      ],
      "focus": [
        "full-stack website reconstruction from browser exploration, without the source code",
        "visual appearance and functional behavior of the rebuilt site"
      ],
      "notes": "Agents run through selectable Harbor agents, so observations and actions depend on the agent. Each task requests 8 CPUs, 8 GiB of memory and 20 GiB of disk per environment, with up to 16 hours for the agent.",
      "observations": [],
      "setting": "interactive"
    },
    "leaderboard_url": "https://website-bench.com/#results",
    "repository_revision": "78f5558e8197bb91b3bab63428c12b92060ddc6f",
    "repository_tree_sha": "a72e8de335c8f531b655eee7d4f9709dec40b366"
  },
  "environments": [
    "web"
  ],
  "evidence": [
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-11-06T06:21:48Z",
      "role": "primary",
      "sha256": "c79cb821975c34552a46f01349edd0869c2f20c9a85d4b7811155d3fb7f7f0b2",
      "supports": [
        "name",
        "kind",
        "summary",
        "environments",
        "topics",
        "scope",
        "notes",
        "details"
      ],
      "url": "https://raw.githubusercontent.com/TIGER-AI-Lab/WebsiteBench/78f5558e8197bb91b3bab63428c12b92060ddc6f/README.md"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-11-06T06:21:44Z",
      "role": "primary",
      "sha256": "d98bd00fb45abc63df0fddd6d9f9f5e669396cd53c157e5e06d6de9d103adf09",
      "supports": [
        "details.benchmark_protocol"
      ],
      "url": "https://raw.githubusercontent.com/TIGER-AI-Lab/WebsiteBench/78f5558e8197bb91b3bab63428c12b92060ddc6f/docs/tasks.md"
    },
    {
      "method": "license_header",
      "retrieved_at": "2026-11-06T06:22:02Z",
      "role": "primary",
      "sha256": "cda5b289b68650d0643c0ded2d43dc797cb7777c4d943f2b03067250de6d3c26",
      "supports": [
        "licenses.code"
      ],
      "url": "https://raw.githubusercontent.com/TIGER-AI-Lab/WebsiteBench/HEAD/LICENSE"
    }
  ],
  "id": "github:tiger-ai-lab/websitebench",
  "identifiers": {},
  "kind": "benchmark",
  "last_verified_at": "2026-11-06T06:21:48Z",
  "licenses": {
    "code": {
      "basis": "repository LICENSE header; not a weights/data license",
      "declared": "Apache-2.0",
      "retrieved_at": "2026-11-06T06:22:02Z",
      "url": "https://raw.githubusercontent.com/TIGER-AI-Lab/WebsiteBench/HEAD/LICENSE"
    },
    "data": null,
    "document": null,
    "weights": null
  },
  "links": [
    {
      "rel": "primary",
      "url": "https://github.com/TIGER-AI-Lab/WebsiteBench"
    }
  ],
  "name": "WebsiteBench",
  "notes": "Read at commit 78f5558e. The README and docs/tasks.md were reviewed; the 70 tasks and 46,916 hidden cases were checked by summing the docs/tasks.md table. The README's paper badge has no link, so no paper relation is recorded. Apache-2.0 is the README's statement for benchmark code, task definitions and verifier. Leaderboard scores were not checked. Nothing was installed or run.",
  "published_at": null,
  "relations": [],
  "review_status": "curated",
  "scope": "core",
  "summary": "Website reconstruction benchmark: agents explore a production-level site through a browser, rebuild a full-stack version, and are scored by hidden functional and visual test cases across 70 tasks.",
  "task_classes": [
    "R02.04.01.02"
  ],
  "topics": [
    "evaluation",
    "web-navigation",
    "long-horizon",
    "ui-design"
  ],
  "updated_at": null,
  "url": "https://github.com/TIGER-AI-Lab/WebsiteBench"
}