dsh-min
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-min
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-min in DeepSeek Harness to install the plugin; the full source code is available at https://github.com/uckkk/dsh-min.
About this plugin
Many model-inference plugins lean on network requests or heavyweight runtimes, adding deployment friction and debugging overhead. dsh-min tackles that pain point by providing a minimal, self-contained entry point for inference tooling so that in-session model calls stay free of external services or tangled dependency chains.
The design boils down to two keywords: pure Node and zero network dependency. The entire plugin is written in vanilla Node.js with no HTTP client, no third-party networking library, and no background service to launch. Once installed, the plugin registers its inference tools directly into the current session, ready to call on demand with no extra configuration.
It is a strong fit if you are building offline-first, zero-dependency workflows or need inference calls inside constrained environments such as CI pipelines, edge devices, and air-gapped machines. Rather than stacking features, dsh-min keeps the surface area intentionally small and lets you focus on what the model actually needs to do.
Use Cases
- Invoke model-inference tools in an offline or intranet environment
- Register inference tools in CI pipelines without any network access
- Lightweight inference tooling on edge devices or sandboxed runtimes
Best For
- Developers building offline-first workflows
- Engineers deploying models in constrained or air-gapped environments
- Pure Node.js practitioners who prefer zero external dependencies
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.