跳转到主要内容

​
browseruse_bench.utils.cli_utils

Command-line argument parsing and error handling utilities.

​
Import

from browseruse_bench.utils import (
    handle_cli_errors,
    create_run_parser,
    create_eval_parser,
    create_base_agent_parser,
)

​
handle_cli_errors

CLI error handling decorator for unified exception and exit code handling. 
def handle_cli_errors(func: Callable[..., int]) -> Callable[..., None]

​
Usage Example

from browseruse_bench.utils import handle_cli_errors

@handle_cli_errors
def main() -> int:
    # Your logic
    return 0  # Success

if __name__ == "__main__":
    main()

​
Behavior

ConditionExit Code
Normal returnFunction return value
KeyboardInterrupt130
Other exceptions1

​
create_run_parser

Create argument parser for the main run script. 
def create_run_parser() -> argparse.ArgumentParser

​
Included Arguments

ArgumentTypeDefaultDescription
--modechoiceallTest mode: single, first_n, specific, sample_n, by_id, all
--countint1Number of tasks for first_n or sample_n mode
--task-idslist-Task ID list for specific mode
--idstr-Single task ID for by_id mode
--timeoutint-Timeout per task (seconds)
--skip-completedflagFalseSkip completed tasks
--debugflagFalseEnable debug mode
--dry-runflagFalseShow command only, do not execute

​
create_eval_parser

Create argument parser for the evaluation script. 
def create_eval_parser() -> argparse.ArgumentParser

​
Included Arguments

ArgumentTypeDefaultDescription
--modestr-Evaluation mode
--modelstr-Evaluation model
--score-thresholdint-Score threshold (LexBench-Browser=60, others=3)
--num-workerint1Number of worker processes
--api-keystr-API Key
--base-urlstr-API Base URL
--dry-runflagFalseShow command only, do not execute

​
create_base_agent_parser

Create base argument parser for Agent run.py. 
def create_base_agent_parser(
    description: str,
    default_tasks_json: str,
    default_output_dir: str
) -> argparse.ArgumentParser
​
description
str
必填
Parser description
​
default_tasks_json
str
必填
Default tasks JSON file path
​
default_output_dir
str
必填
Default output directory

​
Usage Example

from browseruse_bench.utils import create_base_agent_parser

parser = create_base_agent_parser(
    description="Browser-Use Agent Runner",
    default_tasks_json="data/tasks.json",
    default_output_dir="output/"
)
args = parser.parse_args()