# Is Claude Desktop being redesigned by Claude without human supervision? — retained record

> Retained reddit record from reddit:ClaudeAI.

Canonical: https://highsignal.app/data/records/271aca820624f727

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:ClaudeAI)  published 2026-10-28 20:53 ingested 2026-10-29 02:37

[original source ↗](https://www.reddit.com/r/ClaudeAI/comments/1xi08ho/is_claude_desktop_being_redesigned_by_claude/)

## retained text



```
I'm starting to suspect Anthropic has built the ultimate AI development pipeline:

1. Ask Claude to improve the Desktop UI.  
2. Claude removes a useful feature.  
3. Users complain on Reddit and GitHub.  
4. Ask Claude to fix it.  
5. Claude restores the feature but breaks something else.  
6. Ship to production.  
7. Repeat.

First, we lost convenient project grouping. Then branch and worktree controls became harder to find. Context usage disappeared from its familiar location.

After enough complaints, some of those things finally got fixed.

Great!

Except now the indicator showing which sessions are actively working has apparently been replaced with a barely noticeable gray shimmer on the session title.

I regularly run multiple coding agents in parallel. Knowing which ones are working, finished, or waiting for input is not a decorative feature. It's one of the main reasons I use a desktop GUI in the first place.

And somehow, every UI improvement feels like a trade: one regression fixed, another one introduced.

I'm genuinely starting to wonder whether anyone tests these changes with actual multi-project, multi-agent workflows before shipping them.

Maybe the AI is doing the design, the coding, the testing, and the release management.
```



document fetched 2026-10-29 00:22 · canonical https://reddit.com/r/ClaudeAI/comments/1xi08ho/is_claude_desktop_being_redesigned_by_claude
