Introduction¶
DeepSeek Harness (DSH) uses a plugin-based architecture. In agent development, if you need to run specific logic locally without relying on external APIs, you can achieve this by installing a plugin. dsh-mandarin-grade is a purely Node-based calculation tool implemented in this way.
What It Is¶
dsh-mandarin-grade is a plugin maintained by uckkk that provides Mandarin proficiency grade calculation. It does not depend on a network environment.
Core Functionality¶
The tool registered by the plugin is named mandarin_grade. This tool supports precise calculation of Mandarin grades based on parameters and returns detailed results.
Installation and Dependencies¶
Before installation, ensure the environment meets the following dependencies:
- @deepseek-ai/cordis version >= 4.0.1
- @deepseek-ai/dsh-tools version >= 0.1.0-rc.6
Install using the official command:
dsh plugin add github:uckkk/dsh-mandarin-grade
Usage¶
After installation, you can use it by directly invoking the tool mandarin_grade registered by the plugin in a session.
Applicable Scenarios and Notes¶
This plugin is suitable for scenarios that require local language grade determination. Because the plugin runs with the current DSH process permissions, be sure to check its source code and license (MIT) before installation.
Summary¶
dsh-mandarin-grade provides a solution for Mandarin grade calculation without network dependencies. For more details, refer to the plugin directory or source code repository:
- Plugin Directory
- GitHub Repository