Skip to content
UXClaim
Design Ops

Hyperfocus

ADHD-friendly output formatting for Claude Code with evidence-based cognitive accessibility structures.

View source More in Design Ops

What It Does

Hyperfocus restructures Claude’s output for ADHD-friendly comprehension. Instead of dense paragraphs, it reorganizes responses with clear sections, front-loaded answers, short chunks, and lists. Same technical depth—just structured so your brain can absorb it.

How It Works

Activate with /hyperfocus in Claude Code. Choose from three modes:

  • Clean: Professional, scannable structure with short paragraphs
  • Flow (default): What → Why → How framework with re-entry points
  • Zen: TL;DR first, everything in lists/tables for dense topics

Every response follows evidence-based cognitive accessibility principles: short sentences (under 20 words), logical chunking, visual hierarchy with subheadings every 2-3 paragraphs, and white space to prevent cognitive overload.

Use Cases

  • Code reviews: Clean mode for clear, scannable feedback
  • Learning technical concepts: Flow mode with structured explanations
  • Long sessions: Zen mode with TL;DR summaries for easy re-entry
  • Documentation: Any mode for better information retention

Who Benefits

Primarily practitioners with ADHD, but benefits anyone managing information-heavy work. Researchers, developers, designers, and content creators all benefit from structured output that respects cognitive load.

Frequently asked questions

How do I install and activate Hyperfocus?
Install via Claude Code marketplace with `claude plugin install hyperfocus@hyperfocus`, then activate with `/hyperfocus` in any conversation. Choose your preferred mode (clean, flow, or zen) with `/hyperfocus [mode]`. No code changes required.
What's the difference between the three modes?
Clean mode is shortest and most scannable for quick answers. Flow mode uses What→Why→How structure ideal for learning. Zen mode puts TL;DR at the top with everything in lists, best for dense documentation. Switch anytime mid-conversation.
Will Hyperfocus make responses shorter?
No. Hyperfocus restructures for better readability but maintains technical depth and completeness. It removes filler and improves organization, not length. Same information, better structure.
Can I make Hyperfocus always active?
Yes. Use `/hyperfocus persistent` to auto-activate globally, or `/hyperfocus persistent project` for just your current project. The formatting rules re-inject on every turn without drift.
What research supports these formatting rules?
Every rule comes from peer-reviewed research: W3C COGA accessibility standards, Oxford plain English guidelines, WCAG 2.2 cognitive accessibility, and peer-reviewed ADHD information processing studies (PMC6996017).
Does Hyperfocus affect my code or commits?
No. Hyperfocus only restructures prose responses. Code blocks, commits, and PRs remain untouched. It's a formatting skill, not a code modifier.
Which agents support Hyperfocus?
Fully supported on Claude Code, Codex, and GitHub Copilot (via npx skills). Some older agents (Opencode, omp) may require re-activation each turn due to agent limitations.

Glossary

Chunking
Breaking information into smaller, logically grouped sections. ADHD brains process sequential chunks better than parallel information, reducing cognitive overload.
Front-loading
Placing the key answer or main point at the beginning of a response. Important because ADHD readers may not reach the end of dense text.
Cognitive Accessibility
Designing content to be understandable and usable by people with different cognitive processing styles, including ADHD, dyslexia, and learning disabilities.
Visual Hierarchy
Using formatting (headings, spacing, bold text) to show the relative importance of information and enable quick scanning and re-entry.
Task Abandonment
When users stop engaging with content before completing it. Dense walls of text without white space significantly increase task abandonment risk.

More in Design Ops

All →
Design Ops

Autonomous Development Pipeline

Claude Code skill turning specs into shipped code through adaptive phases with feedforward guides, feedback sensors, and requirement traceability.

0xPuncker
Design Ops

Ibras AI Agent

Claude Code marketplace for building and evaluating AI agents. Install plugins to generate complete Mastra evaluation systems with scorers, datasets, experim...

volfadar
Design Ops

Claude Skills

Production-grade Claude AI skills for stock analysis, prompt engineering, meeting documentation, and UX design with visual-first outputs.

bhrpraju
Design Ops

Propose

Run three parallel design agents with different philosophies to generate ranked approach proposals for any design decision.