[BUG] Dispatch on Windows: VM process killed with SIGTERM after ~3 minutes, causing ConnectionRefused on mobile

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
• Claude Desktop: 1.1.8629.0
• Plan: Pro
• Mobile: iOS

What's Wrong

Dispatch sessions consistently killed with SIGTERM after ~3 minutes.
Mobile app shows "API Error: Unable to connect to API (ConnectionRefused)".

Root cause chain (from logs)

1. Initial failure: EXDEV cross-device rename (D:\Temp -> C:\AppData)
2. Workaround applied: TEMP redirected to C:\Temp, permissions fixed
3. Workspace now starts successfully (CONNECTED + REACHABLE)
4. Dispatch still fails: every session (modest-peaceful-davinci,

resume ID: b186db09-2953-4d9c-b34c-69ed093c4112) spawns cleanly
then receives SIGTERM after ~3 minutes with code=0 (clean kill,
not a crash). 15+ attempts over several hours, 100% failure rate.

Logs

See follow-up comment (too long for issue body).

What Should Happen?

Expected

Dispatch session remains active and processes tasks.d remain active and process tasks without timing out.

Error Messages/Logs

Steps to Reproduce

Steps to Reproduce

1. Windows 11 Pro, Claude Desktop 1.1.8629.0
2. Open Cowork > Dispatch on desktop
3. Open Dispatch on iOS mobile app
4. Send any message from mobile
5. Wait ~3 minutes
6. ConnectionRefused error appears on mobile

Claude Model

None

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.1.84 (Claude Code)

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

PowerShell

Additional Information

_No response_