When developing agents in DeepSeek Harness (DSH) that need to handle financial calculation logic, accessing the network directly is often not the optimal solution. Local computation is both stable and efficient. The dsh-gongjijin plugin is designed for this exact purpose, encapsulating housing provident fund loan monthly repayment calculation logic in a pure Node environment.

Plugin Overview

This is a DSH plugin focused on calculating monthly repayments for housing provident fund loans. It is maintained by developer uckkk and released under the MIT license. It does not depend on network requests; all calculations are completed locally, making it suitable for integration into agents that need financial calculation capabilities.

Core Features

The plugin provides the following capabilities:

  1. Housing provident fund loan monthly repayment calculation: supports precise calculation based on parameters and returns detailed results.
  2. Pure Node implementation: no network dependency, with fast and stable execution.

Installation and Activation

Before installation, make sure your environment satisfies the following peerDependencies:

  • @deepseek-ai/cordis (^4.0.1)
  • @deepseek-ai/dsh-tools (>=0.1.0-rc.6)

Install the plugin using the following command:

dsh plugin add github:uckkk/dsh-gongjijin

After installation, the plugin registers the corresponding tools in the session, and they can be called directly.

Typical Usage

After installation, simply invoke the tools registered by this plugin in the session.

Applicable Scenarios and Notes

  • Applicable scenarios: Suitable for agent development that needs to handle business logic related to housing provident fund loans.
  • Permissions: This plugin runs with the permissions of the current DSH process.
  • Security advice: Before installation, be sure to inspect the source code and license. This plugin is licensed under MIT.

Conclusion

dsh-gongjijin addresses the need to query housing provident fund loan monthly repayment amounts through local computation, delivering precise results without a network connection. For more details, please refer to the Plugin Directory or the Source Repository.