Preface¶
Under the plugin-based architecture of DeepSeek Harness (DSH), building localized logic for specific scenarios is an effective way to improve response speed and protect privacy. When an agent needs to handle requests in specific domains, such as skincare advice, calling external APIs directly can often increase latency and is affected by network conditions. The dsh-skin-type plugin aims to address this need by providing offline, local skin-type determination capabilities.
What Is This¶
This is a DSH plugin maintained by uckkk, positioned as an auxiliary tool for model reasoning. It does not rely on network requests; instead, it handles skin-type determination logic through local computation and returns the result in the form of clear recommendations.
Core Features¶
- Skin Type Determination: Provides the
skin_typetool, which receives the relevant parameters and outputs a determination recommendation.
Installation and Enabling¶
Make sure your environment has the necessary dependencies installed, then run the following command:
dsh plugin add github:uckkk/dsh-skin-type
Dependency requirements:
* @deepseek-ai/cordis >= 4.0.1
* @deepseek-ai/dsh-tools >= 0.1.0-rc.6
Typical Usage¶
After installation, no additional configuration is required. In a DSH session, you can use the tool registered by this plugin by calling it directly.
Use Cases and Notes¶
This plugin is suitable for agents that need to integrate skincare recommendation functionality. Note the following:
* It is implemented in pure Node and has no network dependencies at runtime.
* The plugin code runs with the permissions of the current DSH process. Before using it, please verify the security of the source code through the GitHub repository.
* The license is MIT.
Summary¶
By localizing the skin-type determination logic, dsh-skin-type reduces external dependencies and is suitable for scenarios with network environment requirements. For more details, refer to GitHub or the SkillHub directory.