dsh-variance
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-variance
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-variance in your DeepSeek Harness terminal to install this plugin, available at https://github.com/uckkk/dsh-variance .
About this plugin
Variance is one of the most basic yet most frequently overlooked statistics in model-inference and data-analysis workflows. dsh-variance turns that small need into a ready-to-use session tool: implemented in pure Node.js with zero network requests, so you install it once and call it directly from a conversation without any formula boilerplate or external service round-trips.
The plugin is deliberately lean. It ships under the MIT license, spins up with no extra processes or HTTP calls, and is inherently compatible with offline setups and restricted intranet environments. Once registered as a session tool, a single mention in dialogue is all it takes to get a local computation and an instant result back.
If you are a data engineer who sprinkles statistical sanity checks across inference pipelines, or a platform developer who wants to keep every dependency audit-friendly and network-free, dsh-variance offers a tiny, clean entry point for variance calculation — one fewer dependency to install, one fewer container to run, plain and simple.
Use Cases
- Quickly validate output distributions in inference pipelines
- Compute statistics with zero network calls in offline or intranet setups
- One mention in dialogue replaces formula boilerplate or external services
Best For
- Data engineers who need a lightweight stats utility
- Platform developers working in restricted intranet environments
- Inference teams that value zero-dependency, zero-deployment tooling
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.