dsh-knife-skill
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-knife-skill
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-knife-skill in the DeepSeek Harness terminal to install the plugin; the source repository is available at https://github.com/uckkk/dsh-knife-skill
About this plugin
Knife work is the most understated yet most consequential step in home cooking. How thick to cut the pork shreds, whether to score the tomatoes first, if to smash the garlic slices before mincing - these small judgments shape the final dish, yet they leave many home cooks standing at the cutting board with no clear answer. dsh-knife-skill turns that moment of uncertainty into a single parameter-driven tool call that returns a tailored cutting and prep plan for the specific home-style recipe you are preparing.
The plugin registers one tool, knife_skill, which accepts your parameters and hands back a step-by-step knife-work and preparation guide covering common home-cooked dishes. The entire implementation is pure Node.js with zero network dependencies, so it runs fully offline, stays fast, and slots cleanly into any existing conversation or automation pipeline without extra overhead.
It is built for home cooks who want a quick, authoritative cutting reference mid-session, for developers assembling kitchen-workflow automations, and for anyone who prefers to get a standard prep spec without opening a browser tab. Install it once, call it whenever a recipe needs a knife plan, and spend your energy on the cooking itself.
Use Cases
- Look up the cutting steps for a dish before cooking
- Grab a standard prep spec with a one-line question in chat
- Call knife-work parameters inside a kitchen automation workflow
Best For
- Home cooks who want an instant knife-work reference
- Developers building kitchen workflow automations
- Users who prefer offline, zero-dependency prep plans
Related Plugins
whale-girl is a desktop pet in the DSH Web GUI (bottom-right for instant companionship) — draggable, feedable, playable; completed tasks, sessions and companionship time accrue into seniority levels, titles, and memories.
A WebGPU ocean skin for DeepSeek Harness with five-wave Gerstner waves, translucent theme, day/night cycle, and quick controls, installable as plugin or extension.
A Galgame-style conversation view and visual scene editor for DSH web sessions, with a real-time editing mode.
A native skinning plugin for DeepSeek Harness: 8 designer themes, wallpapers, accent colors, and shareable theme packs, built purely on design tokens.