AI Agent Hub
Back to plugins
🖥️

dsh-skill-cockpit

Client Updated 2026.08.17

Run the following command in DeepSeek Harness:

dsh plugin install Itailang2333/dsh-skill-cockpit

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

Run dsh plugin install Itailang2333/dsh-skill-cockpit in the DeepSeek Harness terminal; the source lives at https://github.com/Itailang2333/dsh-skill-cockpit — after install, restart DSH, hard-refresh the browser (Ctrl+Shift+R), then open Settings to find the Skill Cockpit panel.

About this plugin

Skills keep piling up: English names blur together, you cannot tell which ones never fire, and new ideas may overlap with what you already have. dsh-skill-cockpit adds a unified cockpit panel inside DSH Web Settings where you can view, annotate, enable, disable, delete, and import skills on a single page.

Key capabilities include a three-zone view (system, project, pool); a one-line Chinese note per skill that joins search results; host-side usage tracking that records invocation count and last-used time so zombie skills surface at a glance; and a test-fire bench that scores a free-form sentence locally and returns the top-five candidates with matched tokens, letting you check trigger-word overlaps before authoring a new skill.

Best for DSH users who have accumulated multiple Agent Skills and want to reduce management overhead with a fully local, zero-dependency, zero-network setup. All data lives in local JSON files and the plugin coexists cleanly with other skill managers.

Use Cases

  • Browse all installed skills in one settings panel and add a one-line Chinese note for easier recall
  • Spot zombie skills that have never fired by reviewing per-skill invocation counts
  • Run a local test-fire check before authoring a new skill to detect trigger-word overlaps with existing ones

Best For

  • DSH users who have accumulated multiple Agent Skills and want a single dashboard to manage them
  • Developers who prefer a fully local, zero-network, zero-dependency toolchain
  • Chinese-speaking developers who rely on short local notes to keep track of what each skill does