Files
sci-gui-agent-benchmark/evaluation_examples/examples/os/4127319a-8b79-4410-b58a-7a151e15f3d7.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

62 lines
1.4 KiB
JSON

{
"id": "4127319a-8b79-4410-b58a-7a151e15f3d7",
"snapshot": "os",
"instruction": "Use terminal command to count all the lines of all php files in current directory recursively, show the result on the terminal",
"source": "NL2Bash",
"config": [
{
"type": "download",
"parameters": {
"files": [
{
"url": "https://huggingface.co/datasets/xlangai/ubuntu_osworld_file_cache/resolve/main/os/4127319a-8b79-4410-b58a-7a151e15f3d7/setup.sh",
"path": "setup.sh"
}
]
}
},
{
"type": "execute",
"parameters": {
"command": "chmod +x setup.sh",
"shell": true
}
},
{
"type": "execute",
"parameters": {
"command": "bash ./setup.sh",
"shell": true
}
},
{
"type": "execute",
"parameters": {
"command": "export DBUS_SESSION_BUS_ADDRESS='unix:path=/run/user/1000/bus'\nxdg-open /home/user/project",
"shell": true
}
}
],
"trajectory": "trajectories/",
"related_apps": [
"os"
],
"evaluator": {
"func": "check_include_exclude",
"result": {
"type": "vm_terminal_output"
},
"expected": {
"type": "rule",
"rules": {
"include": [
"54"
],
"exclude": []
}
}
},
"proxy": false,
"fixed_ip": false,
"possibility_of_env_change": "low"
}