MSAgent-Bench的benchmark有没有具体的评估脚本？

[YinSonglin1997]

Initial Checks

• I have searched GitHub for a duplicate issue and I'm sure this is something new
• I have read and followed the docs & demos and still think this is a bug
• I am confident that the issue is with modelscope-agent (not my code, or another library in the ecosystem)

What happened + What you expected to happen

modelscope中MSAgent-Bench-中文Agent数据集提到，实验评估包含四个维度：
1.插件调用的准确率：识别api_name后面的是否正确，
2.插件url的准确率：url的地址是否正确
3.插件传入参数的准确率：parameters对应的参数是否正确
4.插件整体的准确率：生成的 function calling是否完全正确，整个json可以被load的格式
请问具体如何实现呢？有没有脚本参考？

Versions / Dependencies

modelscope 1.12.0
modelscope-agent 0.4.1
modelscope_studio 0.3.0
Python 3.10.14

Reproduction script

Issue Severity

None