AI Agent Hub
Back to plugins
dsh-scheduled-send preview

dsh-scheduled-send

Client Updated 2026.09.13

Run the following command in DeepSeek Harness:

dsh plugin install OTTTTTO/dsh-scheduled-send

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

Run dsh plugin install OTTTTTO/dsh-scheduled-send in your terminal to install. Full source at https://github.com/OTTTTTO/dsh-scheduled-send

About this plugin

Sometimes you want to say something later: a quick note for a colleague, a reminder for yourself, or a long context block you want the model to see a few hours from now. dsh-scheduled-send adds a clock button to the DSH Web composer. Draft your message, pick a time, the input clears and the task is persisted. At the set moment the message lands in the bound conversation as a normal user bubble and triggers the model response, indistinguishable from one you typed by hand.

Scheduling runs inside the DSH host process; the browser is just a display layer. You can close the tab or even shut the laptop after setting the time, and delivery still happens as long as the dsh process is alive. If power drops mid-session, the queue restores on restart and re-delivers to the original conversation. Tasks are isolated per conversation, with a sidebar panel for cross-conversation viewing, navigation, and cancellation, plus an inline collapsible list above the input. Mobile touch targets are sized for fingers.

Built for developers who live in DSH Web long conversations, want off-peak questioning, or simply want a more reliable way to handle delayed messages without a browser alarm. Pure JS, zero runtime dependencies, no telemetry, task data stays on local disk.

Screenshots

Use Cases

  • Compose a long prompt and schedule it for delivery hours later so the model picks it up fresh
  • Queue a delayed reminder for an on-call teammate without interrupting them right now
  • Batch-schedule pending messages across conversations and manage them all from one sidebar panel

Best For

  • Developers who live in DSH Web long conversations and prefer off-peak questioning
  • Automation pipelines that need reliable delayed delivery without keeping a browser tab open
  • Minimalist plugin users who value local-only data, zero telemetry, and zero runtime dependencies