dsh-mode
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-mode
Paste the following prompt into your AI chat to install this plugin:
Install the plugin by running dsh plugin install uckkk/dsh-mode in the DeepSeek Harness terminal; the source repository is at https://github.com/uckkk/dsh-mode .
About this plugin
dsh-mode is a lightweight model-inference plugin written entirely in Node.js with zero network dependencies. It targets scenarios where inference must run offline-in sandboxes, air-gapped machines, or CI pipelines-without spinning up external services or proxy layers.
Integration is minimal: once installed, the plugin registers tools that you invoke directly within a session. There is no server to configure, no additional runtime to manage, and no outbound connection required, keeping the dependency footprint as small as possible.
It suits developers who need local, self-contained inference, teams working inside network-restricted containers, or anyone who values a zero-dependency approach. Released under the MIT license, it is free to use, modify, and redistribute.
Use Cases
- Run model inference in fully offline or air-gapped environments
- Invoke inference tools inside network-restricted sandboxes or containers
- Embed zero-external-dependency lightweight inference into CI pipelines
Best For
- Solo developers who need fully local model inference
- DevOps and platform teams running inference inside restricted containers
- Developers who prefer zero network dependencies and minimal runtime overhead
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.