Files
sci-gui-agent-benchmark/evaluation_examples/examples/multi_apps/f918266a-b3e0-4914-865d-4faa564f1aef.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

50 lines
1.5 KiB
JSON

{
"id": "f918266a-b3e0-4914-865d-4faa564f1aef",
"snapshot": "vscode",
"instruction": "Please complete the code and retrieve the output from the Python script 'calculator.py' located on the desktop and save it as 'log.txt' in the same directory as the Python file.",
"source": "GAIA",
"config": [
{
"type": "download",
"parameters": {
"files": [
{
"url": "https://huggingface.co/datasets/xlangai/ubuntu_osworld_file_cache/resolve/main/multi_apps/f918266a-b3e0-4914-865d-4faa564f1aef/calculator.zip",
"path": "/home/user/Desktop/calculator.zip"
}
]
}
},
{
"type": "execute",
"parameters": {
"command": [
"/bin/bash",
"-c",
"unzip /home/user/Desktop/calculator.zip -d /home/user/Desktop/ && rm -rf /home/user/Desktop/calculator.zip"
]
}
}
],
"trajectory": "trajectories/f918266a-b3e0-4914-865d-4faa564f1aef",
"related_apps": [
"vscode",
"os"
],
"evaluator": {
"func": "compare_text_file",
"expected": {
"type": "cloud_file",
"path": "https://huggingface.co/datasets/xlangai/ubuntu_osworld_file_cache/resolve/main/multi_apps/f918266a-b3e0-4914-865d-4faa564f1aef/log_Gold.txt",
"dest": "log_Gold.txt"
},
"result": {
"type": "vm_file",
"path": "/home/user/Desktop/log.txt",
"dest": "log.txt"
}
},
"proxy": false,
"fixed_ip": false,
"possibility_of_env_change": "low"
}