AI Agent Hub
Back to plugins
🤖

dsh-curium

Model Inference Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install uckkk/dsh-curium

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install uckkk/dsh-curium in the DeepSeek Harness terminal to install this plugin; the source repository is available at https://github.com/uckkk/dsh-curium.

About this plugin

Most model-inference toolchains rely on remote APIs or external network calls, making them useless in air-gapped, offline, or strictly isolated network environments. dsh-curium keeps the entire inference pipeline inside the Node.js runtime: from input to output there is zero network dependency.

Named after the element curium, it embodies self-containment. All inference logic executes within a single local process, with no native add-ons, no outbound HTTP calls, and no external services required. A pure Node.js implementation means one runtime, one process, and a minimal footprint.

It is aimed at developers who need fully offline inference, AI applications deployed inside isolated or air-gapped networks, and teams that want to minimise their supply-chain surface. If you need a lightweight local inference entry point without standing up an entire backend, dsh-curium is built for exactly that use case.

Use Cases

  • Run model inference in air-gapped or fully offline environments
  • Execute no-network inference checks inside CI/CD pipelines
  • Deploy a lightweight inference pipeline on isolated hardware

Best For

  • Developers who require fully offline inference capability
  • Small teams shipping AI applications in isolated networks
  • Engineering teams aiming to reduce external dependencies and supply-chain risk