dsh-combination
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-combination
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-combination in the DeepSeek Harness terminal to install this plugin; the full source code is at https://github.com/uckkk/dsh-combination
About this plugin
Combination counting, C(n, k), looks trivial in isolation but quietly becomes a source of off-by-one bugs and inconsistent boundary handling when it is copied across sampling scripts, feature pipelines, and inference pre-processing. dsh-combination folds that one calculation into a single, predictable function call so you can stop re-deriving edge cases and focus on the modelling decisions that actually matter.
The entire implementation is written in plain Node.js with zero network calls and no heavy math libraries on the dependency tree. It runs offline, boots instantly, and fits naturally into air-gapped CI runners, sandboxed containers, or any environment where pulling remote packages is undesirable or outright blocked.
Whether you are assembling a dsh plugin pipeline for model inference, prototyping selection-based sampling strategies, or simply want a drop-in combination utility that works the moment it is installed, dsh-combination gives you a small, auditable, and dependency-free tool with no configuration to fuss over.
Use Cases
- Computing feature combination counts in an inference pipeline
- Verifying combinatorial math in offline CI or air-gapped runs
- Estimating sampling space size during rapid prototyping
Best For
- Developers assembling dsh inference plugin pipelines
- Researchers needing combinatorics in air-gapped settings
- Engineers who prefer zero-config, dependency-free tooling
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.