browseruse_bench/

└── utils/                # Utility modules
│   ├── task_utils.py     # Task processing
│   ├── eval_utils.py     # Evaluation tools
│   ├── cli_utils.py      # CLI argument parsing
│   ├── stats_utils.py    # Statistics tools
│   └── config_loader.py  # Configuration loading

# Import utility functions
from browseruse_bench.utils import (
    REPO_ROOT,
    load_tasks,
    filter_tasks,
    EvaluationModel,
    load_evaluation_model,
    create_run_parser,
    create_eval_parser,
    generate_evaluation_summary,
    load_config_file,
)