trace_synthesis/tasks/506.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

36 lines
1021 B
JSON

{
"sites": [
"shopping"
],
"task_id": 506,
"require_login": true,
"storage_state": "./.auth/shopping_state.json",
"start_url": "http://localhost:28082",
"geolocation": null,
"intent_template": "Buy the highest rated product from the {{product_category}} category within a budget {{dollar_value}}.",
"instantiation_dict": {
"product_category": "meat substitute",
"dollar_value": "between 100 and 200"
},
"intent": "Buy the highest rated product from the meat substitute category within a budget between 100 and 200.",
"require_reset": false,
"eval": {
"eval_types": [
"program_html"
],
"reference_answers": null,
"reference_url": "",
"program_html": [
{
"url": "func:shopping_get_latest_order_url()",
"locator": "document.querySelector(\".order-details-items.ordered\").outerText",
"required_contents": {
"must_include": [
"B01CTR3DLE"
]
}
}
]
},
"intent_template_id": 172
}