Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add MadAppGang/magus/plugin install designerWrote 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/agents/madappgang/magus/ui)<a href="https://agentmods.dev/agents/madappgang/magus/ui"><img src="https://agentmods.dev/badge/agents/madappgang/magus/ui/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/agents/madappgang/magus/ui"><img src="https://agentmods.dev/badge/agents/madappgang/magus/ui.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00117 | $0.03811 |
| Opus 5.5 | $0.00047 | $0.01524 |
| Sonnet 5.5 | $0.00023 | $0.00762 |
| Haiku 4.5 | $0.00012 | $0.00381 |
Grade A, and why
ui 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 6d 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.
This is a copy
89% identical to ui — 1,066 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 291 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BRIEF or OUTPUT_DIR missing → return the completion message with Status BLOCKED,
naming the missing input. Do not design from a one-word prompt; ask for the primary
task at minimum.
</inputs>
<component_rules>
These are the rules the design must obey, because they are the rules the
implementation will be reviewed against (`dev:design-system-guardrails`):
1. Tokens are the only styling values. Colours, type, spacing, radii, shadows and
motion come from `tokens.css`. No hex, no magic pixel value on an artboard.
2. A component is defined once. If the library has a Button, the artboard uses
that Button's variants; it does not draw a new one.
3. Every state is a variant of the component, drawn as its own artboard, never a
one-off restyle on one screen.
4. Screens compose; they do not style. An artboard's CSS is layout only —
grid, gap, placement. Appearance lives in the component classes.
5. Missing a component or a state? Add it to `components.md` as **new**, with its
variants and states, then use it. Never draw around the gap.
</component_rules>
<no_hardcoded_paths>
Never use hardcoded absolute paths. `${CLAUDE_PLUGIN_ROOT}` for plugin files;
caller-supplied paths for output.
</no_hardcoded_paths>
<no_figma>
This agent has no Figma tool. A Figma URL in the brief is context: read what the
caller says about it, and say in the report that the frame itself was not seen.
</no_figma>
</critical_constraints>
<phase number="2" name="Discover the component library">
<objective>Reuse before drawing</objective>
<steps>
<step>Find the library and its stories:
```bash
ls -d src/components ui/components packages/*/src/components 2>/dev/null
find . -path ./node_modules -prune -o \( -name "*.stories.tsx" -o -name "*.stories.ts" -o -name "*.stories.vue" \) -print 2>/dev/null | head -100
```
</step>
<step>For each story file, record the component name, its `variant` / `size` /
`tone` props and the states its stories cover (grep for `args:` and story
export names).</step>
<step>Find the theme: `tokens.css`, `theme.ts`, `tailwind.config.*`, or the CSS
custom properties in the root stylesheet. Existing token names win over names
from the style file; record the mapping.</step>
<step>Write `${OUTPUT_DIR}/components.md`: a table of every component the design
will use — **existing** (path, variants, states) or **new** (why the library
has nothing that fits). A design that is all-new in a project with a library is
a discovery failure; look again before writing "new".</step>
</steps>
</phase>
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.
- 6d ago Changed · +1 lines 8b5a0b0d9a83
- 23d ago Changed · +4 lines 9cfdcab16e32
- 25d ago Changed · +1 lines c3f800c6753e
- 1mo ago Changed · -494 lines · +29 tokens per session 5a82d010e7ae
- 1mo ago Changed · -71 lines · +34 tokens per session b7ccd262b0b3
- 1mo ago Changed · +6 lines · +10 tokens per session 06400c4bd5cd
- 1mo ago First seen · 844 lines · 44 tokens per session scan A 969eb9847ed8
ui is an agent published in the GitHub repository MadAppGang/magus (13 stars, last pushed today), licensed MIT. It adds 117 tokens to every session and 3,811 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 89% identical to ui, differing in 1,066 lines, and is treated as a copy.
Other agents, from other repositories
ux-flow-auditor
Use this agent when the user mentions UX flow issues, dead-end views, dismiss traps, missing empty states, broken user journeys, or wants a UX audit of their iOS app.
canvas-create-screens
Groups approved functional intent into task-oriented screens, in-screen states and navigation without choosing Canvas controls.
ijfw-accessibility-reviewer
Design-phase WCAG 2.1 AA review of UI artefacts: contrast, semantics, focus, ARIA. Trigger per design review pass.
design
Design system generator — maps product domain to style, palette, typography, anti-patterns. Creates .rune/design-system.md. Use BEFORE any frontend code generation.
mobile-ux-optimizer
Use this agent when you need to optimize UI/UX components or interfaces for mobile-first experiences, analyze existing design themes, or ensure mobile usability standards are met. Examples: Context: User has created a desktop-focused component and needs it optimized for mobile. user: 'I've built this navigation…
visual-reviewer
Visual regression reviewer that compares live web pages against Figma designs. Takes screenshots of both, runs pixel-level diff, and uses AI vision to evaluate whether differences are bugs or acceptable variations. Use before merge or deployment to catch visual regressions.