[BUG] claude doctor hangs indefinitely, uses 100% CPU, and soft locks terminal session

I have the same problem, and the command fails：
Last login: Thu Aug 28 11:29:37 on ttys002
/etc/zprofile:7: fork failed: resource temporarily unavailable
/Users/bytedance/.zprofile:2: fork failed: resource temporarily unavailable
/etc/zshrc:7: fork failed: resource temporarily unavailable
/Users/bytedance/.oh-my-zsh/oh-my-zsh.sh:43: fork failed: resource temporarily unavailable
nvm_auto:13: fork failed: resource temporarily unavailable
compdef:95: fork failed: resource temporarily unavailable
/Users/bytedance/.zshrc:124: fork failed: resource temporarily unavailable
WK0M2R63WP%

I'm currently experiencing the same issues, and it will crash my entire computer due to insufficient memory.
The quick escape right now is: pkill -f claude (from another tab, as the first one will be busy), hopefully we're getting a fix soon

The same on my side, OS: Mac OS 15.6 on M1 Pro, terminal oh my zsh

Same problem. Just tried to run doctor and had to hard restart my laptop, event mouse cursor was laggy.

"Checking installation status…"

mds_store process raised to hundreds of CPU.

MacOS 15.4.1 (24E263)
1.0.95 (Claude Code)

Have observed the same on both Wndows and Linux. On Linux I noticed a large amount of "which" processes.

Have observed the same on both Wndows and Linux. On Linux I noticed a large amount of "which" processes.

On MacOs it "feels" like overloading filesystem. Maybe it tries to checm content of node_modules somehow and got into a blackhole.

Same issue here.

Same issue with 1.0.96 on macOS 15.6.1 in Terminal.app.

Terminal hangs, shell scripts stop working across the board, there is a process named 1.0.96 using ~100% CPU in Activity Monitor. Computer just becomes slow across the board. Trying to create a new Terminal tab yields:

[forkpty: Resource temporarily unavailable]
[Could not create a new process and open a pseudo-tty.]

Same issue here (same configuration as what @rben01 posted above).

Same issue here.
MacOS 15.4.1 (24E263)
1.0.95 (Claude Code)

same issue here

same issue here
Windows 10
1.0.96

Same here on Linux/NixOS. It seems to get stuck in an infinite loop spawning thousands upon thousands of npm processes:

m0ar      341452 52.2  0.9 33077428 601380 pts/36 Rl+ 09:48   1:02                  |   |   \_ claude
m0ar      341464  0.0  0.0      0     0 pts/36   Z+   09:48   0:00                  |   |       \_ [npm config get ] <defunct>
m0ar      341471  0.0  0.0      0     0 pts/36   Z+   09:48   0:00                  |   |       \_ [npm config get ] <defunct>
m0ar      341472  0.0  0.0      0     0 pts/36   Z+   09:48   0:00                  |   |       \_ [npm config get ] <defunct>
m0ar      341478  0.0  0.0      0     0 pts/36   Z+   09:48   0:00                  |   |       \_ [npm config get ] <defunct>
[ ... thousands of processes removed for brevity ... ]
m0ar      436566  0.0  0.0      0     0 pts/36   R+   09:50   0:00                  |   |       \_ [npm config get ]
m0ar      436581  0.0  0.0 1053340 59152 pts/36  Rl+  09:50   0:00                  |   |       \_ npm
m0ar      436587  0.0  0.0 1053088 58220 pts/36  Rl+  09:50   0:00                  |   |       \_ npm
m0ar      436597  0.0  0.0 1020148 57824 pts/36  Rl+  09:50   0:00                  |   |       \_ npm
m0ar      436605  0.0  0.0 1016736 54216 pts/36  Rl+  09:50   0:00                  |   |       \_ npm

same here, as reported, Ubuntu 20

Same here.
Win11
Claude 1.0.96

experienced the same - seems to be when running the claude doctor command but not when using the /doctor command

EDIT: for me it's the .96 version

This should be resolved in 1.0.98, released today.

Yeah I’m still seeing 2-2.5GB RAM on an M4, per session.

So running 2-3 in parallel is making my machine feel sluggish.

On Wed, Sep 3, 2025 at 6:52 PM Slava Vishnyakov @.***>
wrote:

slava-vishnyakov left a comment (anthropics/claude-code#6705) <https://github.com/anthropics/claude-code/issues/6705#issuecomment-3250207150> 1.0.102 (Claude Code) still eats a ton of resources even on Apple M4 I think it might be "metallic" animation - this thing: image.png (view on web) <https://github.com/user-attachments/assets/e5995d3b-d5e4-4f3f-87f3-f62c00305df1> — Reply to this email directly, view it on GitHub <https://github.com/anthropics/claude-code/issues/6705#issuecomment-3250207150>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQUI5WARMSMGS3XB4FAJYP33Q4TGZAVCNFSM6AAAAACFAAQ3NKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTENJQGIYDOMJVGA> . You are receiving this because you commented.Message ID: @.***>

I have the same issue, running top in wsl shows infinite "npm config get" processes

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.