Introduction

The plugin system of DeepSeek Harness (DSH) allows developers to encapsulate specific capabilities as tools and invoke them directly within conversation flows. When handling emergency situations or when quick access to specific procedures is needed, a lightweight, network-independent local tool is more efficient than web search.

dsh-cpr-guide is exactly this kind of plugin. It organizes the key steps of CPR (cardiopulmonary resuscitation) first aid by scenario, allowing agents to invoke them directly locally.

Plugin Positioning

This plugin is maintained by user uckkk and is a DSH client plugin implemented purely in Node.js. It does not depend on external networks and returns key handling steps by scenario through built-in data structures.

Core Features

The plugin provides a single tool capability:

  1. Scenario-based guidance: Returns key handling steps according to different first aid scenarios.
  2. Local execution: Implemented purely in Node and can be invoked without internet access, ensuring response speed.

Installation and Activation

Run the following command in the terminal to install it:

dsh plugin add github:uckkk/dsh-cpr-guide

After installation is complete, the plugin registers the corresponding tools in the current session.

Typical Usage

After installation, directly invoke the tool registered by the plugin in a DSH session to obtain emergency steps. The tool name is cpr_guide.

Use Cases and Cautions

This plugin is suitable for scenarios where standardized first aid procedures need to be retrieved quickly.

Please note when using it:

  1. Professional training takes precedence: First aid content should be based on professional training; the plugin provides reference steps.
  2. Permissions and security: The plugin runs with the permissions of the current DSH process. Before installation, it is recommended to review the source code and license (MIT).

Conclusion

dsh-cpr-guide reflects the DSH plugin design philosophy of “on-demand invocation, local-first.” By moving key workflows into local tools, it reduces network request overhead. For more plugin information and source code, refer to the Plugin Directory or GitHub Repository.