AI Agent Hub
Back to plugins
🖥️

dsh-i-have-adhd

Client Updated 2026.08.26

Run the following command in DeepSeek Harness:

dsh plugin install NecromanAlbert/dsh-i-have-adhd

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install NecromanAlbert/dsh-i-have-adhd in your terminal, then restart the DeepSeek Harness desktop app to activate the plugin; source code is available at https://github.com/NecromanAlbert/dsh-i-have-adhd

About this plugin

If you have ever stared at a wall of AI text and zoned out before you could extract what to actually do next, this plugin was built for that exact moment. dsh-i-have-adhd injects ADHD-friendly output rules into the global system prompt so every session, from the very first reply, follows a tighter shape: action first, numbered steps, lists capped at five items, a two-minute closing line, and progress restated only as task position rather than identity or topology.

The mechanism is always-on. Unlike opt-in skills that must be loaded by name, this plugin takes effect as soon as you restart the desktop app, and the rules persist across every response until you explicitly say stop adhd mode. The rules are adapted from ayghri/i-have-adhd (MIT) and keep the action-first, less-is-more spirit without resorting to caveman-speak: technical terms stay precise, nothing is dumbed down.

It is aimed at anyone with ADHD, attention-management struggles, or simply a preference for AI output that is shorter, tighter, and immediately actionable. It does not change the model capabilities; it changes the shape and rhythm of what comes back, so you spend two minutes grabbing your next step instead of twenty minutes hunting for it.

Use Cases

  • Tracking multi-step tasks step by step without drowning in verbose AI text
  • Losing focus while reading replies and needing action-first, numbered structure
  • Technical discussions where precision matters but reading load should stay low

Best For

  • People with ADHD or attention-management challenges seeking tighter, more actionable AI output
  • Technical users who prefer action-first, numbered-list answers
  • DeepSeek Harness developers wanting a consistent global output style across sessions