Skip to content
Agentmart
Submit a tool

frontend-design-guardrails

Use when page, component, form, table, dashboard, responsive layout, or UI review work needs product-quality implementation guardrails.

Skill by funky-eyes

IndexedApache-2.0★ 12
Write a review

Chalk Labs

Need help deploying frontend-design-guardrails?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/funky-eyes-frontend-design-guardrails

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when page, component, form, table, dashboard, responsive layout, or UI review work needs product-quality implementation guardrails. DO NOT USE FOR: backend-only changes or scripts with no user-visible surface.

Use cases: design, frontend, backend, data

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 1, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/frontend-design-guardrails

What's inside

Skill name
frontend-design-guardrails
Description
Use when page, component, form, table, dashboard, responsive layout, or UI review work needs product-quality implementation guardrails. DO NOT USE FOR: backend-only changes or scripts with no user-visible surface.
License
Apache-2.0
Instructions
78 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Frontend Design Guardrails

## Goal

Build usable interfaces that match the target repository's existing UI system and can be verified reliably.

## Reference Principles

- Ant Design baseline: optimize for natural interaction, certainty, meaningful task progress, and systems that can grow without visual drift.
- Design patterns beat freestyle UI: prefer established page patterns, reusable components, tokens, layout rules, copy conventions, feedback patterns, and data-entry/data-display patterns.
- Design-system packets beat vague style adjectives: record the minimum explicit tokens and component decisions needed to make the result repeatable.
- Preview culture matters: make the result inspectable through browser evidence, screenshots, or a clear static fallback.

## Rules

- Prefer the repository's existing component library, theme tokens, layout primitives, icon set, routing, and state patterns.
- If the repository uses Ant Design, Ant Design Pro, Pro Components, or compatible tokens, use their components and interaction patterns first; only build custom controls when the library cannot express the workflow.
- For operational tools, favor dense, calm, scannable layouts over marketing-style hero sections.
- Use semantic controls: buttons for commands, inputs for text, selects/menus for option sets, tabs for views, toggles for binary settings.
- Provide loading, empty, error, disabled, success, and partial states when the workflow needs them.
- Avoid magic pixel offsets, absolute-positioned main layouts, clipped text, overlapping controls, and layout shifts.
- Ensure controls have stable accessible labels or text so browser tests can locate them.
- For mobile and desktop, verify that text fits and primary actions remain reachable.
- Do not invent metrics, testimonials, brand claims, screenshots, or business data. Use real data, labelled placeholders, or explicit empty states.
- Avoid generic AI-looking UI: ungrounded purple gradients, decorative blob backgrounds, arbitrary glass cards, over-rounded nested cards, lorem ipsum, and icon-only controls without accessible names.

## Before Implementation

Freeze:

- target user and workflow
- existing UI library/patterns
- page structure
- critical states
- responsive breakpoints
- validation method
- brand source or visual direction
- data density and hierarchy
- accessibility-critical labels and keyboard path

## Design System Packet

For non-trivial UI work, generate a small design-system packet before coding, inspired by UI UX Pro Max and Open Design's active-design-system workflow:

- `product_type`
- `surface`
- `audience`
- `style_direction`
- `brand_source`
- `color_tokens`
- `typography`
- `spacing`
- `components`
- `data_density`
- `states`
- `accessibility_rules`
- `test_plan`
- `anti_patterns`

Store this in the task plan or component notes only when useful; do not create a persistent design-system folder unless the user or spec asks for it.

If the visual direction is underspecified, propose 2-3 concrete directions with tradeoffs before coding. For implementation tasks with an existing product style, derive the packet from current code and screenshots rather than replacing the product identity.

## Craft Review

Before delivery, self-review the UI across five dimensions:

- hierarchy: primary action, data priority, and scan path are clear
- consistency: spacing, typography, components, and feedback match the system
- function: core workflow works with realistic loading/error/empty data
- accessibility: labels, contrast, keyboard path, and focus states are adequate
- implementation: no layout shifts, clipped text, console errors, or endpoint leaks

## Before Completion

Run or request a browser/visual check through `web-experience-audit`, or state why runtime verification is unavailable and provide static fallback evidence.

From funky-eyes/best-copilot, Apache-2.0.

More from funky-eyes/best-copilot

Signals

Stars
12
Forks
1
Open issues
0
Last commit (this folder)
May 14, 2026 (5 months ago)
Last release
Aug 3, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from funky-eyes

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get frontend-design-guardrails running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying frontend-design-guardrails?

CallGet help
Cite this page

frontend-design-guardrails skill. Agent Mart. https://agentmart.dev/skills/funky-eyes-frontend-design-guardrails (last updated 2026-10-24)