e433f35c1f
- 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.
93 lines
2.2 KiB
JSON
93 lines
2.2 KiB
JSON
{
|
|
"id": "37887e8c-da15-4192-923c-08fa390a176d",
|
|
"snapshot": "os",
|
|
"instruction": "Compress all files in the \"/tmp/test_files\" directory tree that were last modified 30 days ago",
|
|
"source": "NL2Bash",
|
|
"config": [
|
|
{
|
|
"type": "download",
|
|
"parameters": {
|
|
"files": [
|
|
{
|
|
"url": "https://huggingface.co/datasets/xlangai/ubuntu_osworld_file_cache/resolve/main/os/37887e8c-da15-4192-923c-08fa390a176d/setup.sh",
|
|
"path": "setup.sh"
|
|
}
|
|
]
|
|
}
|
|
},
|
|
{
|
|
"type": "execute",
|
|
"parameters": {
|
|
"command": "chmod +x setup.sh",
|
|
"shell": true
|
|
}
|
|
},
|
|
{
|
|
"type": "execute",
|
|
"parameters": {
|
|
"command": "./setup.sh",
|
|
"shell": true
|
|
}
|
|
},
|
|
{
|
|
"type": "execute",
|
|
"parameters": {
|
|
"command": [
|
|
"python",
|
|
"-c",
|
|
"import pyautogui; import time; time.sleep(0.5); pyautogui.click({SCREEN_WIDTH_HALF}, {SCREEN_HEIGHT_HALF}); time.sleep(0.5); pyautogui.hotkey('ctrl', 'alt', 't'); time.sleep(0.5)"
|
|
]
|
|
}
|
|
},
|
|
{
|
|
"type": "activate_window",
|
|
"parameters": {
|
|
"window_name": "Terminal"
|
|
}
|
|
}
|
|
],
|
|
"trajectory": "trajectories/",
|
|
"related_apps": [
|
|
"os"
|
|
],
|
|
"evaluator": {
|
|
"postconfig": [
|
|
{
|
|
"type": "download",
|
|
"parameters": {
|
|
"files": [
|
|
{
|
|
"url": "https://huggingface.co/datasets/xlangai/ubuntu_osworld_file_cache/resolve/main/os/37887e8c-da15-4192-923c-08fa390a176d/eval_20250703.sh",
|
|
"path": "eval.sh"
|
|
}
|
|
]
|
|
}
|
|
},
|
|
{
|
|
"type": "execute",
|
|
"parameters": {
|
|
"command": "chmod +x eval.sh",
|
|
"shell": true
|
|
}
|
|
}
|
|
],
|
|
"func": "check_include_exclude",
|
|
"result": {
|
|
"type": "vm_command_line",
|
|
"command": "bash eval.sh",
|
|
"shell": true
|
|
},
|
|
"expected": {
|
|
"type": "rule",
|
|
"rules": {
|
|
"include": [
|
|
"Success: The task was completed correctly."
|
|
],
|
|
"exclude": []
|
|
}
|
|
}
|
|
},
|
|
"proxy": false,
|
|
"fixed_ip": false,
|
|
"possibility_of_env_change": "low"
|
|
} |