Files
sci-gui-agent-benchmark/evaluation_examples/examples/chrome/cabb3bae-cccb-41bd-9f5d-0f3a9fecd825.json
yuanmengqi e433f35c1f feat: standardize configuration fields across all evaluation examples
- Add `fixed_ip` field to all 369 JSON files in examples directory
  - Set to `true` for 8 files listed in google_chrome.json multi_apps
  - Set to `false` for remaining 361 files
- Add `possibility_of_env_change` field to 363 JSON files missing this field
  - Set to "low" for newly added fields
  - Preserve existing values (4 medium, 2 high) for 6 files that already had this field

This ensures consistent configuration schema across all evaluation examples
while maintaining backward compatibility with existing settings.
2025-07-16 13:45:34 +00:00

85 lines
1.8 KiB
JSON

{
"id": "cabb3bae-cccb-41bd-9f5d-0f3a9fecd825",
"snapshot": "chrome",
"instruction": "Browse spider-man toys for kids and sort by lowest price.",
"source": "online_tasks",
"config": [
{
"type": "launch",
"parameters": {
"command": [
"google-chrome",
"--remote-debugging-port=1337"
]
}
},
{
"type": "launch",
"parameters": {
"command": [
"socat",
"tcp-listen:9222,fork",
"tcp:localhost:1337"
]
}
},
{
"type": "chrome_open_tabs",
"parameters": {
"urls_to_open": [
"https://www.kohls.com/"
]
}
},
{
"type": "activate_window",
"parameters": {
"window_name": "Google Chrome"
}
}
],
"trajectory": "trajectories/",
"related_apps": [
"chrome"
],
"evaluator": {
"func": "check_direct_json_object",
"result": {
"type": "active_tab_html_parse",
"goto_prefix": "https://www.",
"category": "class&url",
"class_multiObject_li": {
"pmpSearch_breadcrumb": [
"Spider-Man",
"Toys",
"Kids"
],
"sbSelector": [
"Price Low-High"
]
},
"url_include_expected_multichoice": {
"Spider-Man": "Spider-Man",
"spiderman": "Spider-Man",
"Toys": "Toys",
"Kids": "Kids",
"S=4": "Price Low-High"
}
},
"expected": {
"type": "rule",
"rules": {
"expected": {
"spider-man": true,
"toys": true,
"kids": true,
"price low-high": true,
"is_other_exist": false
}
}
}
},
"proxy": true,
"fixed_ip": false,
"possibility_of_env_change": "low"
}