AI Agent Hub
Back to plugins
dsh-mobile-plugin preview

dsh-mobile-plugin

Client Updated 2026.08.20

Run the following command in DeepSeek Harness:

dsh plugin install april-jk/dsh-mobile-plugin

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install april-jk/dsh-mobile-plugin in your DeepSeek Harness terminal to add this plugin; the source code is available at https://github.com/april-jk/dsh-mobile-plugin.

About this plugin

DeepSeek Harness binds to 127.0.0.1 by default, so the moment you step away from your desk the local Web UI is out of reach. The DSH Mobile Remote plugin solves this with a single outbound WSS connection to a Relay server — no NAT traversal, no public listener, no inbound port to open. A paired phone simply connects to the Relay browser client and gets full access to the local DSH interface.\n\nSecurity is handled end-to-end: HTTP, SSE, and WebSocket payloads are encrypted with AES-256-GCM between the mobile app and the local Companion, while the Relay forwards only sealed frames and never persists request or response bodies. Pairing is done via a QR code in the Web UI settings; the one-time code and encryption key live in the URL fragment and are never sent in the page request. Device credentials are stored locally with owner-only permissions and can be revoked at any time from the settings page or the dsh-mobile CLI.\n\nThis is the go-to plugin if you run DeepSeek Harness locally but frequently need to check a conversation, switch directories, or tweak settings from your phone after leaving your workstation. After a single QR-code pairing in Settings > Remote Access, your iPhone takes over the full interaction while the local DSH instance keeps exactly the same network footprint as before.

Screenshots

Use Cases

  • Check local DSH conversations from your iPhone after leaving the desk
  • Switch working directories or tweak parameters on the go during a trip
  • Operate Harness over a mobile network at a café without exposing any public port

Best For

  • Developers running DeepSeek Harness locally
  • Mobile workers who need on-demand access to AI conversations
  • Security-conscious users who refuse to open public ports