[Bug] TUI renders garbled/scrambled bold text mid-response — letters split with spaces on macOS
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
What's Wrong?
During longer responses containing bold text and emoji (e.g. numbered lists with bold labels), the TUI terminal output becomes garbled mid-response. Bold text words are scrambled — characters from different words are merged together and letters are split apart with spaces between them.
Example of garbled output observed:
YStaging-irmst onstruction atcsession start — ets thn right context
S a s s — s t r c
This happened multiple times in the same session on responses with numbered lists using bold + emoji formatting. The content/logic was correct but the visual rendering was broken.
Steps to Reproduce
- Start a Claude Code session on macOS (CLI terminal)
- 2. Ask a question that produces a longer response with numbered lists using bold text and emoji (e.g. "✅ Item one ...")
- 3. Observe the TUI output mid-stream — text becomes garbled with scrambled words and spaced-out characters
Expected Behavior
Bold/formatted text renders correctly in the terminal output without scrambling.
Actual Behavior
Mid-response, bold text gets garbled — words are split, characters from adjacent words merge, and letters appear spread out with spaces between them.
Environment
- Platform: macOS
- - Shell: zsh
- - - Claude Code CLI (latest version)
- - - - Terminal: default macOS terminal
What Should Happen?
The TUI should render bold text and emoji correctly throughout the entire response. Numbered lists with bold labels should display without any character scrambling, splitting, or merging. The output should match the actual markdown content being streamed.
Error Messages/Logs
Steps to Reproduce
- Start a Claude Code session in a macOS terminal (Terminal.app or iTerm2)
- 2. Send a prompt that will produce a long response with numbered lists using bold text and emoji (e.g. "list the steps for X with bold labels")
- 3. Watch the terminal output as the response streams in
- 4. Observe garbled/scrambled text mid-response — bold words have characters split with spaces, or characters from adjacent words merged together
- 5. The garbling is visible in the live TUI output but the final rendered content may be correct
Note: This occurred multiple times in the same session on responses containing numbered lists with bold + emoji formatting (e.g. ✅ **Item name** — description).
Claude Model
None
Is this a regression?
Yes, this worked in a previous version
Last Working Version
_No response_
Claude Code Version
2.1.97 (Claude Code)
Platform
Anthropic API
Operating System
macOS
Terminal/Shell
Terminal.app (macOS)
Additional Information
_No response_
This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗