Crash when using Japanese text in TodoWrite - Bun Unicode boundary error

Description

Claude Code crashes with a segmentation fault when using Japanese text in the TodoWrite tool. The crash occurs in Bun's Rust core when handling multi-byte Unicode characters.

Environment

• Claude Code version: Latest (as of 2025-12-25)
• Bun version: 1.3.5
• OS: macOS x64 (Darwin 24.4.0)
• Platform: macOS v15.4.1

Steps to Reproduce

1. Use Claude Code with Japanese language
2. Use the TodoWrite tool with Japanese text containing full-width characters
3. Example todo items that were being used:

• "管理者ページのE2Eテスト作成"
• "管理者ページのE2Eテスト作成中"

Error Message

thread '<unnamed>' (2595110) panicked at /rustc/ed61e7d7e242494fb7057f2657300d9e77bb4fcb/library/core/src/str/mod.rs:833:21:
byte index 2 is not a char boundary; it is inside '）' (bytes 0..3) of \`）\`
note: run with \`RUST_BACKTRACE=1\` environment variable to display a backtrace
============================================================
Bun v1.3.5 (1e86cebd) macOS x64
macOS v15.4.1

panic(main thread): Segmentation fault at address 0x0
oh no: Bun has crashed. This indicates a bug in Bun, not your code.

Analysis

The root cause appears to be a Unicode string boundary issue in Bun's Rust core:

• The error specifically mentions \'）'\ (full-width right parenthesis, U+FF09, 3 bytes in UTF-8)
• Bun is trying to access byte index 2, which is in the middle of this 3-byte character
• This suggests improper handling of multi-byte characters when processing Japanese text

Bun Crash Report Link

https://bun.report/1.3.5/m_11e86cebmgEuhogC____________+y68qB_______A2AA

Workaround

Restart Claude Code and continue the session. The crash is intermittent.

Suggestion

This may need to be reported to Bun as well, but filing here since it affects Claude Code users working in Japanese or other multi-byte character languages.