trace_synthesis/tasks/699.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

72 lines
2.1 KiB
JSON

{
"sites": [
"shopping_admin"
],
"task_id": 699,
"require_login": true,
"storage_state": "./.auth/shopping_admin_state.json",
"start_url": "http://localhost:28083/admin",
"geolocation": null,
"intent_template": "Draft a new marketing price rule for {{topic}} that offers {{rule}} for all customers",
"instantiation_dict": {
"topic": "spring sale",
"rule": "a 20 percent discount site-wide"
},
"intent": "Draft a new marketing price rule for spring sale that offers a 20 percent discount site-wide for all customers",
"require_reset": false,
"eval": {
"eval_types": [
"url_match",
"program_html"
],
"reference_answers": null,
"reference_url": "http://localhost:28083/admin/sales_rule/promo_quote",
"program_html": [
{
"url": "last",
"locator": "document.querySelector(\"[name='name'\").value",
"required_contents": {
"must_include": [
"spring sale"
]
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"website_ids\"').selectedIndex",
"required_contents": {
"exact_match": "0"
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"customer_group_ids\"').selectedIndex",
"required_contents": {
"exact_match": "1"
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"simple_action\"').value",
"prep_actions": [
"document.querySelector('[data-index=\"actions\"]').querySelector('.admin__collapsible-title').click()"
],
"required_contents": {
"exact_match": "by_percent"
}
},
{
"url": "last",
"locator": "document.querySelector('[name=\"discount_amount\"').value",
"prep_actions": [
"document.querySelector('[data-index=\"actions\"]').querySelector('.admin__collapsible-title').click()"
],
"required_contents": {
"exact_match": "20"
}
}
],
"url_note": "GOLD in PRED"
},
"intent_template_id": 258
}