dsh-off-peak-schedule-widget
Run the following command in DeepSeek Harness:
dsh plugin install BaiLiang-233/dsh-off-peak-schedule-widget
Paste the following prompt into your AI chat to install this plugin:
In a DeepSeek Harness creative-mode session, send the install prompt to automatically fetch and run this plugin from https://github.com/BaiLiang-233/dsh-off-peak-schedule-widget
About this plugin
DeepSeek Harness bills by usage, and peak-hour rates are noticeably higher than off-peak. Yet a creator's rhythm does not stop just because the price curve spikes. dsh-off-peak-schedule-widget adds an off-peak scheduling layer on top of the input composer: during peak hours, pressing Enter parks your text in a holding queue instead of submitting it; during off-peak hours the queue is delivered in order to the target conversation you have chosen, saving cost without breaking creative flow.
Key capabilities include up to six custom peak windows (midnight-crossing supported), a target conversation that can be pinned to an existing session or created on demand, adjustable workspace concurrency from 1 to 3, drag-to-reorder with force-execute and undo (both double-confirmed), persistent config in runtime-config.json that survives restarts, and a built-in one-click updater plus run log in the panel. During off-peak hours the input pipeline remains fully native; the plugin only adds a sidebar entry and a thin status strip above the input area.
It is built for users who want to batch agent tasks until off-peak, who prefer delivering all output to a single pinned conversation instead of spawning a new one each time, or who need fine-grained control over exactly which time windows freeze their input.
Use Cases
- Draft inputs during peak hours and batch-deliver them to a target conversation during off-peak
- Keep all agent output flowing into one pinned conversation instead of spawning a new one each time
- Freeze submissions across up to six custom peak windows to control exactly when tasks execute
Best For
- Creators who run frequent agent tasks on DeepSeek Harness and want to reduce peak-hour costs
- Users who prefer pinning output to a single conversation rather than creating a new one repeatedly
- Team members who need granular control over submission timing and queue management
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.