dsh-gadolinium
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-gadolinium
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-gadolinium in your terminal to install the plugin. Source repository: https://github.com/uckkk/dsh-gadolinium.
About this plugin
Many model-inference plugins rely on native module compilation or make network calls at runtime to fetch files, which breaks in offline settings, CI sandboxes, or strict-security deployments. dsh-gadolinium takes a simpler path: the entire inference pipeline is implemented in pure Node.js with zero network requests, so it runs locally the moment you install it.
Two capabilities stand out. First, the pure JavaScript runtime eliminates the need for node-gyp or any C++ toolchain, meaning no extra compilation step on Windows, Linux, or macOS. Second, the zero-network design means no external service calls and no weight downloads during inference, making it well-suited for intranet clusters, air-gapped labs, or environments where outbound traffic is locked down.
Who is it for? If you need a minimal-dependency, offline-capable inference component to drop into a Node project, or you are building a CI pipeline that must pass without any external network egress, dsh-gadolinium's light footprint and deterministic behavior fit the bill.
Use Cases
- Running model inference locally in air-gapped labs or intranet clusters
- Executing inference steps in CI pipelines without outbound network access
- Loading inference modules across Windows, Linux, and macOS without a C++ toolchain
Best For
- Engineers deploying inference in offline or strict-security environments
- DevOps engineers building CI/CD pipelines without external network egress
- Node.js full-stack developers wanting zero-compile, minimal-dependency inference integration
Related Plugins
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.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.