Introduction

The core design philosophy of DeepSeek Harness (DSH) is “Everything is a plugin.” When developing agents or orchestrating workflows, if you need to perform asset allocation calculations locally, you typically need to introduce external libraries or call network interfaces. The dsh-asset-split plugin provides a local computation solution that does not rely on the network.

Plugin Positioning

This is a tool-type plugin focused on asset allocation calculations, maintained by uckkk. It is implemented in pure Node.js and does not involve any network requests, making it suitable for use in DSH workflows that require a stable, offline environment.

Core Features

The plugin provides asset allocation calculation capabilities. By calling the tools registered by the plugin during a session, it can perform precise calculations based on the input parameters and return detailed results.

Installation and Dependencies

Before installing, make sure your environment satisfies the following dependency requirements:

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

The installation command is as follows:

dsh plugin add github:uckkk/dsh-asset-split

The plugin uses the ES Module specification, and the type field in package.json is set to module.

Usage

After running the installation command above, the plugin will register the corresponding tools in the current session. You can use the plugin by calling these tools directly in a DSH workflow or conversation.

Notes

  • The plugin runs with the permissions of the current DSH process. Please make sure the source is trustworthy.
  • The source code is hosted on GitHub. It is recommended to review the license (MIT) and source implementation before installing.
  • Dependency versions must match strictly, otherwise the plugin may fail to load properly.

Summary

dsh-asset-split addresses the problem that asset allocation calculations often require external dependencies or network access. It is a lightweight, purely local solution, suitable for scenarios that have high requirements for deterministic calculation results.