[BUG] Cowork: "Claude Code process exited with code 1" on Windows 11 Pro after VM successfully boots (MSIX install)

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?

Environment:

• OS: Windows 11 Pro, Version 10.0.26200, Build 26200, 64-bit
• Claude Desktop: 1.3036.0.0 (MSIX package: Claude_1.3036.0.0_x64__pzs8sxrjxfjjc)
• CPU: Intel Core i9-9980HK
• Machine: Dell Precision 7540
• Hyper-V: Enabled
• WSL: 2.6.3.0 (kernel 6.6.87.2-1)
• Docker Desktop: 29.1.3 (installed, using WSL2 backend)
• Locale: it-IT

What's wrong:

Two stages of failure. First stage resolved by user workaround; second stage persists.

Stage 1 — Original error (RESOLVED by user):
"Impossibile avviare l'area di lavoro di Claude - VM service not running. The service failed to start."

Root cause identified through diagnosis: VirtualizationFirmwareEnabled: False even with Hyper-V enabled in Windows features, because:

1. Intel VT-x was disabled in BIOS firmware (enabled manually)
2. HVCI / Memory Integrity was holding exclusive access to VT-x (disabled via Windows Security → Core Isolation)

After enabling VT-x in BIOS + disabling Memory Integrity + reboot:

• VMMS service: Running ✅
• HVCI Enabled: 0 ✅
• VM now successfully boots

Stage 2 — Current error (NOT RESOLVED):

Opening Cowork now produces: Claude Code process exited with code 1

The VM boots, but the Claude Code process inside the VM crashes immediately with generic exit code 1. This appears to match the pattern reported in issues #37822, #39270, #46029 (Claude Code process spawned by Desktop app crashes instantly with code 1 while the same binary works correctly when invoked directly from terminal).

Claude Code in standalone terminal works perfectly — the issue is specific to how Claude Desktop / Cowork spawns the Claude Code process inside its VM.

Expected behavior:

1. The original "VM service not running" error should detect missing virtualization prerequisites (VT-x, HVCI conflict) and surface actionable guidance instead of a generic error
2. Cowork should not crash with exit code 1 after successful VM boot

UX suggestion:
Add preflight checks at Cowork launch:

• Detect VirtualizationFirmwareEnabled: False and tell users to enable VT-x in BIOS
• Detect active HVCI / Memory Integrity and warn about the conflict with Cowork
• On exit code 1, surface the actual error from inside the VM rather than just the generic wrapper message

Diagnostic data collected:

• Get-CimInstance Win32_Processor .VirtualizationFirmwareEnabled: was False (now True after BIOS fix)
• Get-Service vmms .Status: was empty (now Running)
• Win32_DeviceGuard.VirtualizationBasedSecurityStatus: was 2 (HVCI) (now 2 but services running = 0)
• Win32_DeviceGuard.SecurityServicesRunning: was [2] (now [0])

Happy to provide debug logs from Help → Troubleshooting → Share debug logs if useful.

What Should Happen?

Expected behavior:

1. The original "VM service not running" error should detect missing virtualization prerequisites (VT-x, HVCI conflict) and surface actionable guidance instead of a generic error
2. Cowork should not crash with exit code 1 after successful VM boot

Error Messages/Logs

**Stage 1 — Original error (RESOLVED by user):**
"Impossibile avviare l'area di lavoro di Claude - VM service not running. The service failed to start."

Root cause identified through diagnosis: `VirtualizationFirmwareEnabled: False` even with Hyper-V enabled in Windows features, because:
1. Intel VT-x was disabled in BIOS firmware (enabled manually)
2. HVCI / Memory Integrity was holding exclusive access to VT-x (disabled via Windows Security → Core Isolation)

After enabling VT-x in BIOS + disabling Memory Integrity + reboot:
- `VMMS service: Running` ✅
- `HVCI Enabled: 0` ✅
- VM now successfully boots

**Stage 2 — Current error (NOT RESOLVED):**

Opening Cowork now produces: `Claude Code process exited with code 1`

The VM boots, but the Claude Code process inside the VM crashes immediately with generic exit code 1. This appears to match the pattern reported in issues #37822, #39270, #46029 (Claude Code process spawned by Desktop app crashes instantly with code 1 while the same binary works correctly when invoked directly from terminal).

**Claude Code in standalone terminal works perfectly** — the issue is specific to how Claude Desktop / Cowork spawns the Claude Code process inside its VM.

Steps to Reproduce

Current error (NOT RESOLVED):**

Opening Cowork now produces: Claude Code process exited with code 1

The VM boots, but the Claude Code process inside the VM crashes immediately with generic exit code 1. This appears to match the pattern reported in issues #37822, #39270, #46029 (Claude Code process spawned by Desktop app crashes instantly with code 1 while the same binary works correctly when invoked directly from terminal).

Claude Code in standalone terminal works perfectly — the issue is specific to how Claude Desktop / Cowork spawns the Claude Code process inside its VM.

Claude Model

Not sure / Multiple models

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

Claude Code v2.1.112

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

Windows Terminal

Additional Information

_No response_