Preface

DeepSeek Harness (DSH) adopts an “everything is a plugin” architecture, allowing developers to extend model reasoning capabilities through plugins. In model inference or Agent development, specific domain knowledge or tools sometimes need to be injected into the workflow. dsh-heat-level is such a plugin; it addresses the heat-level judgment problem in cooking scenarios, helping the model output concrete home cooking plans.

Plugin Overview

dsh-heat-level is a plugin maintained by uckkk, categorized under model inference. It focuses on taking parameter inputs and returning corresponding heat-level judgments and cooking plans.

Core Features

  • Heat-Level Judgment: Returns home cooking plans based on input parameters.
  • Pure Local Execution: Implemented with pure Node.js, without network dependencies.

Installation and Activation

Installing this plugin requires a DeepSeek Harness environment. Use the following command to install it:

dsh plugin add github:uckkk/dsh-heat-level

Typical Usage

After installation, no additional configuration is required. You can directly call the tool registered by this plugin in a DSH session. The tool name is heat_level.

Applicable Scenarios and Notes

  • Dependencies: This plugin depends on @deepseek-ai/cordis (^4.0.1) and @deepseek-ai/dsh-tools (>=0.1.0-rc.6).
  • Runtime Permissions: The plugin runs with the permissions of the current dsh process.
  • Security Notice: Since the plugin runs as local code, it is recommended to review the source code before installation and confirm that the MIT license meets project requirements.

This plugin is suitable for scenarios where cooking logic needs to be integrated into the model inference workflow. For more details, refer to DSH plugin directory or GitHub repository.