trace_synthesis/tasks/547.json
yuyr a84d51a101 1. 增加r1生成综合策略代码和输出;
2. 增加tasks;
3. 增加analysis部分,对策略进行归纳分类,然后进行评测。
2025-04-17 17:40:15 +08:00

38 lines
1.0 KiB
JSON

{
"sites": [
"shopping_admin"
],
"task_id": 547,
"require_login": true,
"storage_state": "./.auth/shopping_admin_state.json",
"start_url": "http://localhost:28083/admin",
"geolocation": null,
"intent_template": "Add a new {{option}} option {{value}} to the {{base_setting}} of {{product}}",
"instantiation_dict": {
"option": "color",
"value": "brown",
"base_setting": "size S",
"product": "Phoebe Zipper Sweatshirt"
},
"intent": "Add a new color option brown to the size S of Phoebe Zipper Sweatshirt",
"require_reset": false,
"eval": {
"eval_types": [
"program_html"
],
"reference_answers": null,
"reference_url": "",
"program_html": [
{
"url": "http://localhost:28083/admin/catalog/product/edit/id/1130/",
"locator": "document.querySelector('[data-index=\"configurable\"').outerText",
"required_contents": {
"must_include": [
"Phoebe Zipper Sweatshirt-S-Brown"
]
}
}
]
},
"intent_template_id": 252
}