dsh-minimal-rules
Run the following command in DeepSeek Harness:
dsh plugin install sd1g1/dsh-minimal-rules
Paste the following prompt into your AI chat to install this plugin:
Install the plugin in DeepSeek Harness by running dsh plugin install sd1g1/dsh-minimal-rules. The source code is available at https://github.com/sd1g1/dsh-minimal-rules .
About this plugin
In a DeepSeek Harness model-inference pipeline, overly complex strategy layers can blur the reasoning boundary and make debugging harder. dsh-minimal-rules takes the opposite approach: it applies a deliberately lean set of rules to guide inference, giving the model clear guardrails without sacrificing flexibility.
As a plugin in the model-inference category, its core philosophy is less is more. It inserts the smallest viable rule surface into the reasoning pipeline, adds no architectural overhead, and avoids stacking verbose strategy configurations. It drops into an existing Harness workflow with minimal setup.
It suits developers who want lightweight, rule-based control over the inference stage and prefer a minimal configuration surface over elaborate strategy stacks. If your goal is a clear, predictable reasoning boundary without complexity, dsh-minimal-rules is a sensible starting point.
Use Cases
- When a reasoning pipeline needs a lean set of rules to bound model output
- When you want to cut strategy-layer complexity and reduce multi-layer debugging cost
- When you need a clear, predictable inference boundary without a verbose strategy stack
Best For
- Developers who prefer a minimal configuration surface and embrace the less-is-more philosophy
- Engineers building or maintaining inference pipelines within DeepSeek Harness
- Teams that want lightweight rules in place of heavy strategy layers to lower maintenance cost
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.