dsh-plugin-workbuddy-gateway
Run the following command in DeepSeek Harness:
dsh plugin install Acoder416/dsh-plugin-workbuddy-gateway
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install Acoder416/dsh-plugin-workbuddy-gateway in your terminal to install from https://github.com/Acoder416/dsh-plugin-workbuddy-gateway
About this plugin
DSH expects an OpenAI-compatible endpoint for model inference, but WorkBuddy's subscription API does not expose one out of the box. dsh-plugin-workbuddy-gateway runs a lightweight local gateway that translates WorkBuddy's international or domestic subscription interface into a standard OpenAI-compatible format, letting DSH consume it just like any other third-party model provider.
Everything is managed from a single WorkBuddy panel inside DSH settings: start, stop, or restart the gateway; import desktop credentials or authenticate via browser OAuth; check credit balances and sign-in status; rotate across multiple accounts with per-model rate-limit recovery; switch between the international and domestic regions; and write gateway models into the llm-pi-ai provider config in one click. Rate limits, cooldowns, and unavailability are tracked at the account-and-model level so requests are not wasted before a slot reopens.
The plugin is a fit for DSH users who hold a WorkBuddy subscription and want a unified inference entry point, especially those juggling international and domestic accounts, needing multi-account rotation, or wanting at-a-glance credit visibility. It ships as source code with no plugin-market dependency and runs on Windows, macOS, and Linux.
Use Cases
- Translate WorkBuddy international or domestic subscription endpoints into OpenAI-compatible routes for DSH inference
- Manage credit balances, sign-in status, and per-model rate-limit recovery across multiple accounts from one DSH panel
- Switch between international and domestic regions while the gateway rotates accounts and handles 429 throttling
Best For
- DSH users with a WorkBuddy subscription who want a single inference entry point
- Power users juggling international and domestic accounts who need rotation and credit visibility
- Developers who prefer source installs, local gateways, and no plugin-market dependency
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.