[Bug] Windows 中文系统：1.38.0+ 版本内置 skill 文件编码错误 || [Bug] Windows Chinese system: 1.38.0+ version built-in skill file encoding error

What version of Kimi Code CLI is running?

版本信息：昨天 1.38.0 正常，今天升级1.39后报错

Which open platform/subscription were you using?

/login

Which model were you using?

kimi-k2.6

What platform is your computer?

x64

What issue are you seeing?

错误信息： 'utf-8' codec can't decode byte 0xcf in position 46
系统环境：Windows, utf-8编码

What steps can reproduce the bug?

启动不了

What is the expected behavior?

No response

Additional information

session-c2a41388-903c-4a8a-b378-a4f9e00d1a00.zip 附上report

---

What version of Kimi Code CLI is running?

Version information: 1.38.0 was normal yesterday, but an error occurred after upgrading to 1.39 today.

Which open platform/subscription were you using?

/login

Which model were you using?

kimi-k2.6

What platform is your computer?

x64

What issue are you seeing?

Error message: 'utf-8' codec can't decode byte 0xcf in position 46
System environment: Windows, utf-8 encoding

What steps can reproduce the bug?

Can't start

What is the expected behavior?

No response

Additional information

session-c2a41388-903c-4a8a-b378-a4f9e00d1a00.zip Attached is report

Start by reproducing the startup failure on Windows with a Chinese system setting, using the reported upgrade from 1.38.0 to 1.39 and the UTF-8 decode error. Trace the Kimi Code CLI startup path that loads its built-in skill files and identify the file or encoding boundary involved. Done means the CLI starts successfully on the reported environment without the decoding error.