Panic: UTF-8 byte boundary error when processing Chinese characters

[tytsxai]

Description

Claude Code CLI crashes with a Rust panic when processing output containing Chinese characters.

Error Message

thread '<unnamed>' (693199) panicked at /rustc/ed61e7d7e242494fb7057f2657300d9e77bb4fcb/library/core/src/str/mod.rs:833:21:
byte index 8 is not a char boundary; it is inside '。' (bytes 6..9) of `康）。`
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
fatal runtime error: failed to initiate panic, error 5, aborting
zsh: abort      claude --permission-mode dontAsk

Environment

• Claude Code Version: v2.0.76
• Model: claude-opus-4-5-20251101
• OS: macOS (Darwin 25.1.0)
• Shell: zsh

Steps to Reproduce

1. Start Claude Code with claude --permission-mode dontAsk
2. Ask Claude to perform a task that generates Chinese text output
3. In my case, I was updating documentation files containing Chinese content
4. The crash occurred while Claude was updating docs/architecture.md with Chinese text

Root Cause Analysis

This appears to be a string slicing error in Rust where the code attempts to slice a UTF-8 string at byte index 8, which falls inside the multi-byte Chinese character 。 (Chinese period, occupying bytes 6-9).

The string fragment 康）。 suggests the crash happened while processing Chinese text containing:

• 康 (3 bytes)
• ） (3 bytes)
• 。 (3 bytes)

Expected Behavior

Claude Code should properly handle UTF-8 multi-byte characters and only slice strings at valid character boundaries.

Workaround

Restart Claude Code and retry the operation. The issue may be intermittent.