agentic — Claude Code skill: design system for AI interfaces
A skill that gives the agent a design-system rulebook for conversational AI interfaces: tokens, component states and accessibility checks.
npx skills add bergside/awesome-design-skills --skill agentic -a claude-codeWhat agentic does
This skill turns the assistant into a design-system author. The style it encodes is modern and bold: Playfair Display for body and headings, JetBrains Mono for code, a type scale from 14 to 40, a primary orange of #FF5701 and an 8pt spacing grid. The agent reaches for these values when it proposes tokens and component rules instead of inventing a palette.
The workflow is ordered. Restate the design intent in one sentence, define tokens and constraints, then describe component anatomy, states and behaviour for keyboard, pointer and touch. Accessibility is a hard requirement here: WCAG 2.2 AA, visible focus, 44px+ touch targets and reduced-motion support, each written so it can be tested in code. When looks and accessibility collide, the skill says accessibility wins.
Output follows a fixed outline: context and goals, tokens, component rules, acceptance criteria, anti-patterns and a QA checklist for code review. The catch is scope. You get guidance, not a component library, so copying markup out of it is the wrong move. It also sits badly on a project that already has a design system, because agentic's rules will fight the existing ones.
What it helps with
- write tokens and component states for a conversational AI interface
- turn accessibility guidance into criteria you can test in code
- build a QA checklist for design-system code review
- record anti-patterns and migration notes for legacy components
- set a concise, low-jargon tone for interface copy
Example prompts for the agent
- Write a design-system guideline in the agentic style: tokens, component states and accessibility criteria
- Audit our sign-in screen against the agentic skill and give me a code-review checklist
- Describe the modal's anatomy and states following agentic, including anti-patterns
How to install agentic
1Skills CLI (Vercel)
npx skills add bergside/awesome-design-skills --skill agentic -a claude-code2Manually: put the files in place
~/.claude/skills/agentic/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/agentic/
1Skills CLI (Vercel)
npx skills add bergside/awesome-design-skills --skill agentic -a codex2Manually: put the files in place
.agents/skills/agentic/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add bergside/awesome-design-skills --skill agentic -a cursor2Manually: put the files in place
.cursor/skills/agentic/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install bergside/awesome-design-skills agentic2Manually: put the files in place
.github/skills/agentic/SKILL.md1Skills CLI (Vercel)
npx skills add bergside/awesome-design-skills --skill agentic -a gemini-cli2Platform CLI
gemini skills install https://github.com/bergside/awesome-design-skills --path skills/agenticManually: put the files in place
.devin/skills/agentic/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add bergside/awesome-design-skills --skill agentic -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about agentic
How do I install agentic in Claude?
Run npx skills add bergside/awesome-design-skills --skill agentic -a claude-code. The skill lands at ~/.claude/skills/agentic/SKILL.md.
What does the agentic skill do?
It guides the agent when writing design-system guidance: tokens, typography, component states and testable accessibility requirements.
Which platforms support agentic?
There are install commands for Claude, Codex, Cursor, GitHub Copilot, Gemini CLI and OpenCode. GitHub Copilot uses gh skill install bergside/awesome-design-skills agentic.
Source text
SKILL.md · MIT
--- name: agentic description: Conversational AI-first interface with minimal controls, clear outcomes, and delegated task flows for agentic workflows. license: MIT metadata: author: typeui.sh --- <!-- TYPEUI_SH_MANAGED_START --> # Agentic Design System Skill (Universal) ## Mission You are an expert design-system guideline author for Agentic. Create practical, implementation-ready guidance that can be directly used by engineers and designers. ## Brand The agentic design style emphasizes conversational interactions, clear outcomes, and minimal controls, allowing users to delegate tasks to AI instead of manually managing complex workflows. ## Style Foundations - Visual style: modern, bold - Typography scale: 14/16/18/24/32/40 | Fonts: primary=Playfair Display, display=Playfair Display, mono=JetBrains Mono | weights=100, 200, 300, 400, 500, 600, 700, 800, 900 - Color palette: surface/subtle layers | Tokens: primary=#FF5701, secondary=#F6F6F1, success=#16A34A, warning=#D97706, danger=#DC2626, surface=#FFFFFF, text=#111827 - Spacing scale: 8pt baseline grid ## Accessibility WCAG 2.2 AA, keyboard-first interactions, visible focus states, semantic HTML before ARIA, screen-reader tested labels, reduced-motion support, 44px+ touch targets, high-contrast support ## Writing Tone concise, confident, helpful, clear, friendly, professional, action-oriented, low-jargon ## Rules: Do - prefer semantic tokens over raw values - preserve visual hierarchy - keep interaction states explicit - design for empty/loading/error states - ensure responsive behavior by default - document accessibility rationale ## Rules: Don't - avoid low contrast text - avoid inconsistent spacing rhythm - avoid decorative motion without purpose - avoid ambiguous labels - avoid mixing multiple visual metaphors - avoid inaccessible hit areas ## Expected Behavior - Follow the foundations first, then component consistency. - When uncertain, prioritize accessibility and clarity over novelty. - Provide concrete defaults and explain trade-offs when alternatives are possible. - Keep guidance opinionated, concise, and implementation-focused. ## Guideline Authoring Workflow 1. Restate the design intent in one sentence before proposing rules. 2. Define tokens and foundational constraints before component-level guidance. 3. Specify component anatomy, states, variants, and interaction behavior. 4. Include accessibility acceptance criteria and content-writing expectations. 5. Add anti-patterns and migration notes for existing inconsistent UI. 6. End with a QA checklist that can be executed in code review. ## Required Output Structure When generating design-system guidance, use this structure: - Context and goals - Design tokens and foundations - Component-level rules (anatomy, variants, states, responsive behavior) - Accessibility requirements and testable acceptance criteria - Content and tone standards with examples - Anti-patterns and prohibited implementations - QA checklist ## Component Rule Expectations - Define required states: default, hover, focus-visible, active, disabled, loading, error (as relevant). - Describe interaction behavior for keyboard, pointer, and touch. - State spacing, typography, and color-token usage explicitly. - Include responsive behavior and edge cases (long labels, empty states, overflow). ## Quality Gates - No rule should depend on ambiguous adjectives alone; anchor each rule to a token, threshold, or example. - Every accessibility statement must be testable in implementation. - Prefer system consistency over one-off local optimizations. - Flag conflicts between aesthetics and accessibility, then prioritize accessibility. ## Example Constraint Language - Use "must" for non-negotiable rules and "should" for recommendations. - Pair every do-rule with at least one concrete don't-example. - If introducing a new pattern, include migration guidance for existing components. <!-- TYPEUI_SH_MANAGED_END -->
© typeui.sh · MIT · bergside/awesome-design-skills