corrinehu/dsh-workbuddy-connect

将 WorkBuddy 桌面 App 包含的模型自动接入 DeepSeek Harness,零配置使用。Bring the models in the WorkBuddy desktop app into DeepSeek Harness with zero configuration.

What it solves

This project provides a plugin for DeepSeek Harness (DSH) that allows users to automatically integrate and use models from the WorkBuddy and WorkBuddy AI desktop applications without manual configuration. It bridges the gap between the WorkBuddy ecosystem and the DSH interface, enabling access to models like GLM-5.3, DeepSeek-V4, and Kimi-K3 directly within DSH.

How it works

The plugin leverages the login status and credentials of the installed WorkBuddy desktop apps (both domestic and international versions). It automatically detects the installed apps, extracts the necessary authentication tokens, and maps the available models and their associated pricing (credits) to the DSH model selector. It supports image inputs for compatible models and handles reasoning effort levels (low/high/max) by dynamically detecting available tiers for specific models.

Who it’s for

Users who have WorkBuddy or WorkBuddy AI desktop applications installed and want to use those models within the DeepSeek Harness (DSH) environment across Web, Desktop, or TUI interfaces.

Highlights

  • Zero Configuration: Automatically connects to WorkBuddy apps upon installation and activation.
  • Dual Version Support: Simultaneously supports both domestic WorkBuddy and international WorkBuddy AI, keeping accounts and credits separate.
  • Credit Monitoring: Displays real-time remaining credits for the active account directly in the DSH input box.
  • Dynamic Reasoning Tiers: Supports manual detection of reasoning effort levels for models that don't explicitly declare them.
  • Model Management: Allows users to toggle the visibility of specific models in the selector and customize context window settings.
  • Multimodal Support: Enables image pasting and dragging for models that support vision capabilities.

Related

  • Project
  • Project
  • Project
  • Project