Asynchronous development with Claude Code on a phone
A developer set up a system to remotely launch Claude Code from a mobile device using a combination of WireGuard, mosh, and tmux. The architecture uses a home server as a jump box, allowing connection to a work PC without installing VPN clients on the corporate network.
Notifications are handled by a self-hosted ntfy service, integrated through Claude Code webhooks. The system automatically sends push messages when the model requests clarification or completes a task. Mosh ensures a stable connection when switching between Wi-Fi and cellular networks, while tmux preserves the session when the terminal is closed.
This approach enables running long tasks, putting the phone down, and receiving notifications only when intervention is required.
Related stories
- 57% of public Claude Code subagents inherit Bash
- Toolog keeps a forensic log of Claude Code tool calls
- Stemma compiles agent rules into CLAUDE.md and AGENTS.md
- Spotify's shunt blocks Claude Code reads over 350 lines
- Context Engineering Kit adds judge agents to Claude Code
- ccswitch juggles Claude Code accounts from one config
Comments
No comments yet. Be the first.
Join the conversation
Sign in with Google to leave a comment. Your name and avatar come from your Google profile, and the comment appears after moderation.