,___,
(o,o)
/) )
--"-"--
Design × AI resources for the people designing and building agentic experiences — open source, inclusive by default, built in public.
→ agent-design-primer — The reference for agentic UX: trust, refusals, plan visibility, accessibility. Five layers, 30,000 words, free.
→ ai-design-field-guide — 46 ways to work with AI design tools. The thinking applies regardless of which tool you have open.
→ designer-skills — 87 agentic skills and 27 commands across 8 plugins for Claude Code and Gemini CLI. Research, systems, UI, interaction, and delivery.
→ ai-design-skills — 44 skills for designing AI products specifically: alignment reasoning, evaluation, agent orchestration, prompt architecture.
→ designpowers — A team of 10 design agents. Discovery through retrospective. You direct.
→ agent-ready — A skill + Figma plugin that scores how AI-readable your design files are, and compensates for the gaps.
→ perceptual-protocols — Nine protocols for telling an agent how the work should feel. mood.md for what it should feel like, vocab.md for what the words mean, voice.md for how it writes, tokens.md for what the design system means, and five more. One repo, one install.
mood-protocol, voice-protocol, and tokens-protocol used to live in their own repositories. They're folders in perceptual-protocols now — the old repos are archived and still resolve.
→ inclusive-design-skills — Accessibility skills for AI coding agents.
→ agent-ready — The shared/checks.js module implements all 13 agent-readiness checks as pure functions. Fork it, add checks, plug into your pipeline.
→ ai-design-skills — Skill files are plain markdown with structured frontmatter. One file per skill, one commit to add one.
→ REFERENCES.md — Full bibliography: CHI papers, alignment research, NNGroup, Google PAIR. Every source that informed the skills collection.
→ RESEARCH.md — Research-to-practice mappings: for each paper, what was adopted, what was reframed, and how the academic concept became the practitioner skill.
→ agent-design-primer — The agency spectrum, user mental models, and trust calibration chapter are grounded in HCI literature and designed to be citable.
Owl-Listener is MC Dean. The work here is about what happens when design meets AI and the discipline of designing the relationship between a human and a new kind of intelligence.
Everything is MIT or CC-licensed. Fork it, adapt it, ship it.