Panic on Korean string handling - byte index not char boundary

Environment

• Claude Code version: 2.0.76
• OS: Rocky Linux 9
• Platform: linux x86_64

Error Message

thread '<unnamed>' (1097796) panicked at /rustc/ed61e7d7e242494fb7057f2657300d9e77bb4fcb/library/core/src/str/mod.rs:833:21:
byte index 5 is not a char boundary; it is inside '리' (bytes 3..6) of `고리 `
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
fatal runtime error: failed to initiate panic, error 5, aborting
Aborted (core dumped)

Context

• Working on a Korean help system with diagrams
• The string 고리 appears to be part of a word like "카테고리" (category)
• Claude Code crashed completely with core dump

Possible Cause

• UTF-8 byte boundary not properly handled when truncating Korean strings
• String slicing during markdown/diagram rendering

Additional Notes

• Korean characters use 3 bytes in UTF-8
• The panic occurs at byte index 5, which is inside the character '리' (bytes 3-6)
• This suggests the code is using byte indexing instead of character-aware indexing