[BUG]'s 💔 I love Claude Code, but the pace is outrunning the quality - and it's costing us trust

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?

Claude Code is an incredible tool that many of us have come to rely on daily. However, there's a growing pattern that's becoming increasingly frustrating:

Releases ship faster than bugs get fixed.

We're seeing a cycle where:

• Critical functionality breaks (current example: LSP has been broken since 2.0.67)
• Multiple GitHub issues get filed by frustrated users
• New versions release with new features built on top of broken foundations
• No official acknowledgment of known issues
• Users spend hours troubleshooting problems that turn out to be known bugs

The LSP situation is just the latest example, but it's part of a broader pattern. Scrolling through GitHub issues reveals thread after thread of users experiencing the same problems, often without any official response.

The real cost isn't just the bugs themselves - it's the time we lose:

• Hours debugging setups that were never going to work
• Searching through issues to confirm "it's not just me"
• The back-and-forth of hoping each new release fixes things, only to find it doesn't
• Reading release notes as a survival mechanism before daring to update

When your tool becomes essential infrastructure for developers, stability isn't optional - it's expected.

What Should Happen?

1. Acknowledgment of known issues

When bugs are reported by multiple users, a simple "we're aware and working on it" would save countless hours of user frustration and duplicate debugging.

2. A Known Issues page or Status dashboard

An official place where users can quickly check if something is a known problem before spending time troubleshooting. This is standard practice for developer tools at this level.

3. Quality over velocity

Please consider slowing the release cadence if it means shipping more stable builds. We'd rather wait for something that works than receive frequent updates that introduce or perpetuate issues.

4. Fix before feature

Critical bugs in existing functionality should take priority over new features. Shipping new LSP features while core LSP is broken sends the wrong message.

5. Better communication

Even a monthly "state of Claude Code" update would help. What's being worked on? What's known to be broken? What's the priority? Transparency builds trust.

---

I'm writing this because i want Claude Code to succeed. It's genuinely one of the best tools out there when it works. Please help me keep believing in it.

With hope for better times,
DKMaker 🛠️

Error Messages/Logs

Steps to Reproduce

This isn't a traditional bug with reproduction steps - it's a pattern. But here's how any user can experience it:

1. Encounter a problem with Claude Code (e.g., LSP not working, or any other recurring issue)
2. Assume it's your setup - spend 30-60 minutes debugging, reinstalling, checking configurations
3. Search GitHub issues - find 5-15 other users reporting the same problem, some weeks or months old
4. Notice the silence - no official acknowledgment, no "known issue" label, no ETA
5. Check release notes for the latest version - see new features, but no mention of your bug
6. Update anyway, hoping it's secretly fixed - it isn't
7. Repeat with the next release

Alternative reproduction path:

1. Find a working Claude Code feature you rely on
2. Update to a new version
3. Discover the feature is now broken
4. Check GitHub - others have the same issue
5. Wait for fix while new features keep shipping
6. Lose trust in the update process

Time to reproduce: Unfortunately, just a normal week of using Claude Code

Claude Model

None

Is this a regression?

Yes, this worked in a previous version

Last Working Version

2.0.67

Claude Code Version

2.0.76

Platform

Anthropic API

Operating System

Ubuntu/Debian Linux

Terminal/Shell

Other

Additional Information

I'm writing this late December 2025, a bit like a wish for 2026.

Claude Code is my primary development tool - I'm not looking for alternatives, and I don't want to. I just want to be able to trust it.

My hope for 2026 is simple: a bit more stability, a bit more communication, and the confidence that updating won't break my workflow.

Here's to a great 2026. 🥂

DKMaker