AI Agent Hub
Back to plugins
⚙️

dsh-host-guide

Workflow Updated 2026.08.21

Run the following command in DeepSeek Harness:

dsh plugin install uckkk/dsh-host-guide

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

Install it in DeepSeek Harness by running dsh plugin install uckkk/dsh-host-guide; the full source repository is at https://github.com/uckkk/dsh-host-guide .

About this plugin

Hosting a dinner at home rarely feels as effortless as it looks. Balancing the menu, arranging seating, pacing the courses, remembering the small courtesies that make guests feel welcome—all of it adds up and is easy to fumble on the spot. dsh-host-guide turns home-dinner hospitality etiquette into a single callable Agent tool, so you can pass in your gathering parameters in a conversation and walk away with a complete, structured hosting plan instead of scrolling through forum posts or relying on fuzzy memory.

The entire plugin revolves around one tool, host_guide. You supply inputs such as guest count and cuisine preference, and it returns a full home-dinner hosting plan covering every beat from greeting guests at the door to seeing them out. Under the hood it is a pure Node.js implementation with zero network calls: it runs fully offline, stays lightweight, and produces deterministic output every time you invoke it.

It is a good fit for anyone weaving everyday household decisions into an Agent workflow. Whether you are building a home-assistant chatbot, prototyping a family planning agent, or simply want to ask in a chat,

Use Cases

  • Pass in guest count and cuisine preference in a chat to get an instant home-dinner hosting plan
  • Embed a ready-made hospitality tool into a household-assistant Agent workflow
  • Generate welcoming, seating, and send-off suggestions for tonight's dinner without any web lookup

Best For

  • Developers building household-assistant or lifestyle-decision Agent workflows
  • Home cooks who want to turn hosting intuition into a reusable, structured reference
  • Users who prefer lightweight, fully offline tools with zero network calls