trace_synthesis/tasks/695.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

87 lines
2.4 KiB
JSON

{
"sites": [
"shopping_admin"
],
"task_id": 695,
"require_login": true,
"storage_state": "./.auth/shopping_admin_state.json",
"start_url": "http://localhost:28083/admin",
"geolocation": null,
"intent_template": "Add a simple product named {{product}} with {{stock}} in stock, available in size {{size}} and color {{color}}, priced at ${{price}}",
"instantiation_dict": {
"product": "Energy-Bulk Man Yoga Pant",
"stock": "50",
"size": "38",
"color": "yellow",
"price": "69.99"
},
"intent": "Add a simple product named Energy-Bulk Man Yoga Pant with 50 in stock, available in size 38 and color yellow, priced at $69.99",
"require_reset": false,
"eval": {
"eval_types": [
"url_match",
"program_html"
],
"reference_answers": null,
"reference_url": "http://localhost:28083/admin/catalog/product",
"program_html": [
{
"url": "last",
"locator": "document.querySelector('[name=\"product[price]\"').value",
"required_contents": {
"exact_match": "69.99"
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"product[name]\"').value",
"required_contents": {
"must_include": [
"Energy-Bulk Man Yoga Pant"
]
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"product[quantity_and_stock_status][qty]\"').value",
"required_contents": {
"exact_match": "50"
}
},
{
"url": "last",
"locator": "document.querySelector('[data-role=\"selected-option\"').outerText",
"required_contents": {
"must_include": [
"bottom"
]
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"product[size]\"').value",
"required_contents": {
"exact_match": "179"
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"product[color]\"').value",
"required_contents": {
"exact_match": "60"
}
},
{
"url": "last",
"locator": "document.querySelector('[data-index=\"category_ids\"').outerText",
"required_contents": {
"must_include": [
"bottoms"
]
}
}
],
"url_note": "GOLD in PRED"
},
"intent_template_id": 256
}