trace_synthesis/tasks/590.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

57 lines
1.5 KiB
JSON

{
"sites": [
"gitlab"
],
"task_id": 590,
"require_login": true,
"storage_state": "./.auth/gitlab_state.json",
"start_url": "http://localhost:28084/primer/design",
"geolocation": null,
"intent_template": "Create a milestone for the upcoming {{event}} starting on {{start_date}} and ending on {{end_date}}",
"instantiation_dict": {
"event": "event of product launch",
"start_date": "1/16/2023",
"end_date": "1/30/2023"
},
"intent": "Create a milestone for the upcoming event of product launch starting on 1/16/2023 and ending on 1/30/2023",
"require_reset": false,
"eval": {
"eval_types": [
"url_match",
"program_html"
],
"reference_answers": null,
"reference_url": "http://localhost:28084/primer/design/-/milestones",
"program_html": [
{
"url": "last",
"locator": "document.querySelector(\"#content-body\").outerText",
"required_contents": {
"must_include": [
"product launch"
]
}
},
{
"url": "last",
"locator": "document.querySelector('.block.start_date').outerText",
"required_contents": {
"must_include": [
"Jan 16, 2030"
]
}
},
{
"url": "last",
"locator": "document.querySelector('.block.due_date').outerText",
"required_contents": {
"must_include": [
"Jan 30, 2030"
]
}
}
],
"url_note": "GOLD in PRED"
},
"intent_template_id": 339
}