[BUG]

Resolved 💬 5 comments Opened Apr 23, 2026 by ajeevbhatia-gif Closed Apr 24, 2026

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?

Summary

Claude Dispatch has been non-functional on my mobile device since the Anthropic outage on Wednesday, April 15, 2026 (5 days ago). Dispatch was working normally that morning until approximately 10:00 AM, when the outage began. After Anthropic services came back online, Dispatch never recovered for my account. I am on an active Max plan, emailed support, received an unhelpful response, escalated to a live agent, and have not received a follow-up.

I attempted the fix from #45937 and it did not resolve the issue. Diagnostic evidence below strongly suggests this is a server-side pairing state that only Anthropic can reset.

Environment

  • Plan: Max (active, billed through Anthropic)
  • Account email: ajeevbhatia@gmail.com (same account on both devices)
  • Computer: MacBook, macOS Tahoe 26.2
  • Claude Desktop version: 1.3883.0 (auto-updated from 1.3561.0 during troubleshooting)
  • Phone: iPhone 16 Pro
  • Claude mobile app version: 1.260416.0

Expected behavior

Per the Help Center article on assigning tasks from anywhere in Cowork, Dispatch should be accessible from the mobile app when paired with a running Claude Desktop session on a Max plan. It was working normally for me until April 15.

Actual behavior

In the mobile app, when I open Dispatch:

  • The screen displays: "Can't reach your desktop. Check that your computer is online and desktop app is open."
  • The status icon under the Dispatch header shows "Asleep"
  • This persists regardless of desktop state. Claude Desktop is running and in the foreground, the Mac is awake and plugged in, same Wi-Fi, same account.

Timeline

  • Wed, April 15, 2026, ~10:00 AM: Using Dispatch normally
  • Wed, April 15, 2026: Anthropic outage begins
  • After outage resolved: All other Claude services came back online; Dispatch never recovered on my account
  • 5 days since the incident: Dispatch has remained in "Asleep" / "can't reach your desktop" state. Support emailed, escalated to a live agent, no resolution.

Troubleshooting already attempted

  • Confirmed Max subscription is active
  • Updated Claude Desktop to the latest version (1.3883.0), fully quit and relaunched
  • Updated Claude mobile app from the App Store (1.260416.0), force-quit and relaunched
  • Logged out and back in on both devices
  • Verified both devices are on the same account and same Wi-Fi network
  • VPN disabled, firewall not blocking
  • Claude Desktop running and foregrounded, Mac awake, not on battery saver

None of the above resolved the issue.

Attempted fix from #45937 — did not resolve

Followed the exact fix from https://github.com/anthropics/claude-code/issues/45937#issuecomment-4221855762:

  1. Quit Claude Desktop (verified via pgrep that process fully exited)
  2. rm -rf ~/Library/Application\ Support/Claude/local-agent-mode-sessions/
  3. rm ~/Library/Application\ Support/Claude/bridge-state.json
  4. Reopened Claude Desktop (auto-updated to 1.3883.0 in the process)

Result: Dispatch still shows "Asleep" / "Can't reach your desktop" on mobile.

Diagnostic: bridge-state.json has not been regenerated, even after Desktop has been running for a while and the mobile app has attempted to reconnect by force-quitting and reopening. Desktop is otherwise healthy and writing plenty of other state files on relaunch (config.json, cowork-enabled-cli-ops.json, Preferences, Cookies, Network Persistent State, fcache, window-state.json, extensions-blocklist.json, DIPS, and others touched in the same session). But specifically it does not attempt to create a new bridge session.

Interpretation: Unlike #45937 where local bridge state was corrupted and regenerated cleanly on relaunch, in my case Desktop appears to receive a backend signal that no pairing exists, so it never attempts to bridge at all. This suggests my account's server-side Dispatch pairing state was wiped or disabled when the April 15 incident was marked resolved, and has not been restored. There is no user-facing Dispatch toggle in Claude Desktop Settings to manually trigger re-pairing.

Related issues

This matches the pattern in #40958 (Dispatch non-functional after a "Resolved" incident, Max plan, no support response) and #49545 (Dispatch showing desktop offline for multiple weeks). Appears to be a category of backend pairing state that did not recover for affected accounts when the April incidents were closed.

What I need from Anthropic

Re-enable or re-initialize the Dispatch pairing state on my account server-side. No amount of client-side troubleshooting will resolve this.

Impact

I use Claude every day as a core part of my workflow and Dispatch is a primary reason I am on the Max tier. Losing a paid feature for 5 days with no response from support is a significant problem for a daily paying customer.

Happy to provide account ID, request IDs, or logs on request.

What Should Happen?

Dispatch should connect

Error Messages/Logs

Steps to Reproduce

Open dispatch on mobile

Claude Model

None

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

1.3883.0

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

Terminal.app (macOS)

Additional Information

_No response_

View original on GitHub ↗

This issue has 5 comments on GitHub. Read the full discussion on GitHub ↗