AI Agent Hub
Back to plugins
🖥️

deepseek-plugin-liangwengu

Client Updated 2026.09.03

Run the following command in DeepSeek Harness:

dsh plugin install VinciBeans/deepseek-plugin-liangwengu

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

Run dsh plugin install VinciBeans/deepseek-plugin-liangwengu in your DeepSeek Harness terminal to install the plugin; the source is hosted at https://github.com/VinciBeans/deepseek-plugin-liangwengu .

About this plugin

Pushing heavy generations, code tasks, or multi-step reasoning through DeepSeek Harness means the difference between a fast response and a long queue often comes down to one thing: whether you hit a peak or an off-peak compute window. Memorising the rule—weekdays 09:00-12:00 and 14:00-18:00 are Liang Wengfeng (peak), everything else plus the entire weekend is Liang Wengu (off-peak)—requires a mental lookup before every single prompt. deepseek-plugin-liangwengu replaces that lookup with a permanent, at-a-glance badge right in your Harness session bar, so you never have to dig through docs or do arithmetic in your head.

The badge updates in real time with a per-second countdown that auto-flips the instant the window changes. Cross-day valleys running from Friday 18:00 straight through to Monday 09:00 can stretch up to 63 hours; once a valley crosses the 24-hour mark the UI switches to whole-day units so you never mistake a multi-day window for a short one. All timing is computed in Asia/Shanghai regardless of where your browser is, and the colour scheme follows your DSH theme with automatic dark-mode adaptation, blending into the interface without requiring any extra setup.

If you regularly push long generations, code debugging, or reasoning chains through Harness, or you simply want to build the habit of scheduling heavy work for off-peak hours and keeping peak slots for quick queries, this plugin removes the guesswork entirely. One badge, two seconds, and you instantly know who to call and how long the window lasts.

Use Cases

  • Checking whether you are in an off-peak window before launching a long generation
  • Planning batch inference jobs during the Friday-to-Monday cross-day valley
  • Keeping the badge visually consistent when toggling DSH dark mode

Best For

  • Users who regularly push long generations or code tasks through Harness
  • Developers who want to build the habit of scheduling heavy work for off-peak hours
  • Teams that need a single Asia/Shanghai reference for compute windows