dsh-native-reasoning-slider
Run the following command in DeepSeek Harness:
dsh plugin install WSL043/dsh-native-reasoning-slider
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install WSL043/dsh-native-reasoning-slider in your terminal or inside DSH to install the plugin; source is available at https://github.com/WSL043/dsh-native-reasoning-slider.
About this plugin
Different models expose different reasoning-intensity tiers in DSH, and cycling through the stock selector can feel cramped and repetitive. dsh-native-reasoning-slider distills the control surface into two compact capsules above the input box: a Model capsule for switching and an Intensity capsule that expands into a tight inline slider on tap, so you adjust reasoning power without leaving the input area or opening a separate panel.
The plugin holds one hard rule: it renders only the tiers the current model actually publishes, never inventing or padding options. While you drag, the change is a purely local preview; the single commit to DSH fires only after you release. Three visual modes are available. Official mode restores the stock DSH selector in full. Native mode keeps a quiet capsule-plus-popup-slider layout with no animation. Energy mode triggers a hand-rolled WebGL particle burst and soft glow on each commit: still at Off or mid tiers, a gentle continuous burn at the maximum tier, and full unloading of the renderer the moment the popup closes. Colour presets cover light and dark themes, can be saved globally or per model, and legacy two-colour configs auto-migrate with a recommended track colour. Keyboard navigation is complete throughout. On privacy, the plugin reads no credentials, collects no telemetry, and makes no network requests of its own.
It suits everyday DSH users who want a direct, low-noise way to tune reasoning intensity without stacking several overlapping plugins. One caveat: the project was archived and set to read-only on 31 August 2026, and current DSH releases already ship native reasoning-tier controls in the model menu, so new users may prefer the official widget.
Screenshots
Use Cases
- Tune the current model reasoning-intensity tier right inside the DSH input area
- Save per-model light or dark colour presets for a consistent visual style
- Flip between the quiet Native slider and the particle-burst Energy mode on demand
Best For
- Everyday DSH developers who prefer in-place reasoning-intensity adjustment
- Minimalist users who want one clean plugin instead of several overlapping ones
- Power users who enjoy custom palettes and full keyboard navigation
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.