UI Skills is a collection of skills and lessons for design engineers working on user interfaces. Developers and coding agents use it through a command-line interface or MCP tools to browse and retrieve design-engineering guidance. The catalogue add-ons contain selected UI skills and an instruction for using them.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ibelick/ui-skills --skill component-canvasgit clone --depth 1 https://github.com/ibelick/ui-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ibelick/ui-skills/component-canvas)<a href="https://agentmods.dev/skills/ibelick/ui-skills/component-canvas"><img src="https://agentmods.dev/badge/skills/ibelick/ui-skills/component-canvas/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ibelick/ui-skills/component-canvas"><img src="https://agentmods.dev/badge/skills/ibelick/ui-skills/component-canvas.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- Socket pass
- Snyk pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00024 | $0.01751 |
| Opus 5.5 | $0.00010 | $0.00700 |
| Sonnet 5.5 | $0.00005 | $0.00350 |
| Haiku 4.5 | $0.00002 | $0.00175 |
Grade A, and why
component-canvas scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Component canvas
Build a canvas that shows the project's existing components as they really behave. The canvas wraps components. It never changes them.
Ground rules
- Do not modify component source, exports, styles, hooks, or behavior to make the canvas work. Components must not import or depend on canvas code.
- Do not patch libraries, browser APIs, rendered component DOM, or portal roots.
- Show only the components the user asked for. Adding components, installing dependencies, changing a component's design, or making the canvas public requires permission first.
- Keep existing access controls.
- Leave unrelated work alone. If removing an earlier canvas integration, undo only its changes to components.
- Follow any restrictions on commands, browser access, dependencies, and edits.
Before building
Read the repository conventions, routes, component public APIs, design tokens, and test setup. Use the existing framework and libraries.
Ask the user only when the answer changes what gets built or touches shared code: which components are in scope, whether a new dependency is acceptable, or whether a live service may be previewed. For everything else, choose a sensible default, state it in the report, and proceed.
Build order
- Add a minimal canvas route or entry point with one group and one component. Confirm it works.
- Add the remaining groups with sample data.
- Add variant comparisons, then advanced gestures and persistence, one feature at a time.
Keep canvas code in one place. Check imports and package exports when moving files. Follow the project's language, validation, and test conventions. Do not add architectural layers or directory names the project does not already use.
Keep layout and camera logic separate from sample data and component source.
Layout
- Fill the viewport with a plain background.
- If the project already supports light and dark themes, add an accessible toggle using its existing theme API. Apply the selected theme to the canvas and native portals. If theme support is absent, omit the toggle. Do not invent themes or change component styles.
- Give every group the same padding, radius, and thin dashed border, subtle but visible in both themes. Use an existing muted color at roughly 25% opacity for borders. Place its plain title 8px above the border, using the muted text color without opacity.
- Use wide groups with equal-width example slots filling each row. Wrap when needed and keep generous horizontal and vertical gaps within and between groups. Spread the layout without enlarging components.
- Groups cannot be dragged. Their positions need not be hard-coded. Measure their actual bounds, including title, border, and padding. Ignore rendering-containment placeholders (empty boxes that reserve space before content renders), since they do not reflect real content size.
- Omit background dots, decorative cards, page menus, instruction banners, selection tools, and block editing. Component cards and menus may appear as examples.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 124 lines · 24 tokens per session scan A ef2d4b55a2ab
component-canvas is a skill published in the GitHub repository ibelick/ui-skills (10,113 stars, last pushed 4d ago), licensed MIT. It adds 24 tokens to every session and 1,751 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-14.
Other skills, from other repositories
frontend-design
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.
display-glasses-with-jetpack-compose-glimmer
Provides guidelines for developing projected Android XR apps for display glasses using the Jetpack Compose Glimmer UI toolkit. Covers Glimmer design principles, workflows, and interaction models for the glasses form factor. Use this skill to build an Android XR Augmented Experience app with Jetpack Compose Glimmer.…
glance-canonical-layouts
Use this skill to create home screen widgets using Jetpack Glance Canonical Layouts from the Android Platform Samples project, or when asked to implement standard widget layout patterns (including Check List, Image Grid, Snap Scrolling, and Standard Toolbar).
read
Fetches URLs and PDFs, then summarizes or returns clean Markdown; can also search the web through DeepSeek when built-in search is thin. Use when asked to read, fetch, quote, cite, convert, or save a URL or PDF. Not for local text files already in the repo.
visual-edit
Open and collaboratively edit a running local app in Design, with shared fallback previews and source handoff. Use when the user asks to inspect, share, or edit a real local app in Design.
visual-plan
Turn ordinary text plans into rich interactive visual plans with diagrams, file maps, annotated code, open questions, and UI/prototype review when useful.