dsh-llm-workbuddy
Run the following command in DeepSeek Harness:
dsh plugin install Axiaohungry/dsh-llm-workbuddy
Paste the following prompt into your AI chat to install this plugin:
Install the plugin in an existing DeepSeek Harness environment by running dsh plugin install Axiaohungry/dsh-llm-workbuddy (source: https://github.com/Axiaohungry/dsh-llm-workbuddy), then restart DSH to use it from the settings panel.
About this plugin
If you are running DSH and want to reach WorkBuddy China model services without a dedicated adapter, dsh-llm-workbuddy closes that gap. It lets developers integrate, authenticate, and manage models entirely within the DSH WebUI, with no WorkBuddy CLI required.
Two independent authentication modes are at the core: a WorkBuddy API key (environment variable or multiple DSH-stored keys with a dropdown switcher) and a China-site token login (multi-account persistence, real-time credit balance, and daily request tracking). Model management is equally straightforward. Pull the live model catalogue for the current account, then edit ID, display name, context window, max output tokens, and thinking level per model, or add, remove, and restore entries against the built-in directory.
It is built for developers who already use DSH and need WorkBuddy China inference, especially in team settings where different API keys or enterprise token accounts must coexist. The plugin is MIT-licensed; after installation and a DSH restart, everything is configured from the settings panel.
Use Cases
- Quickly integrate WorkBuddy China model services in DSH
- Team members sharing different API keys or enterprise token accounts
- Configure models via WebUI without installing WorkBuddy CLI
Best For
- DSH developers wanting to integrate WorkBuddy inference services
- Teams managing multiple model credentials and accounts
- Engineers preferring WebUI configuration over CLI workflows
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.