Skip to content
Agentmart
Submit a tool

frontend-design

Design or reshape frontend UI with an explicit visual direction, stronger aesthetic judgment, and browser-verifiable outcomes instead of generic AI-looking…

Skill by gclm

IndexedMIT
Write a review

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/gclm-frontend-design

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

Design or reshape frontend UI with an explicit visual direction, stronger aesthetic judgment, and browser-verifiable outcomes instead of generic AI-looking layouts.

Use cases: design, frontend

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. Checked Oct 25, 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

What's inside

Skill name
frontend-design
Description
Design or reshape frontend UI with an explicit visual direction, stronger aesthetic judgment, and browser-verifiable outcomes instead of generic AI-looking layouts.
License
MIT
Instructions
66 lines in SKILL.md · view on GitHub

Files (2)

  • SKILL.md
  • frontend-task-shaping.md · references/

SKILL.md (first lines)

# Frontend Design

Use this skill when the task is not just frontend cleanup, but choosing or executing a meaningful UI direction.

## Read First

- `docs/overview/roadmap.md` and `harness/state.md` if they exist
- `harness/history.md` if it exists
- `docs/reference/conventions.md` if it exists
- the closest UI files, design notes, screenshots, or reference examples
- `references/frontend-task-shaping.md` when the task could drift between marketing, product-flow, redesign, or prototype work
- `skills/planning/brainstorm/references/reference-reuse-boundary.md` when the task involves borrowing from another product

## Goals

- choose an intentional visual direction before implementation
- avoid generic AI-looking layout, spacing, typography, and color choices
- make visual trade-offs explicit instead of hiding behind vague “modern UI” language
- preserve usability while increasing distinctiveness and coherence
- leave behind UI that can be verified with browser evidence, not only by reading code
- keep design exploration aligned to the actual product surface instead of drifting into the wrong artifact type

## Design Rules

- start by naming the visual direction in plain language
- define the intended feel before choosing components
- make typography deliberate: hierarchy, rhythm, density, and personality should be chosen, not inherited by default
- make color deliberate: define a clear palette and contrast model instead of falling back to generic neutrals
- make spacing and composition deliberate: avoid evenly padded card grids unless they truly fit the product
- use motion only when it clarifies focus, state change, or spatial transition
- backgrounds should contribute to atmosphere or hierarchy, not default to flat emptiness
- when using a reference product, prefer reusing information architecture and interaction structure over copying low-quality implementation details
- when the design must stay within an existing system, preserve that system instead of forcing novelty

## Process

1. Restate the UI problem in terms of user perception and product surface, not only component structure.
2. Use `references/frontend-task-shaping.md` to identify the artifact type, real page coverage, and any fork/upstream boundary that should shape the work.
3. Surface any hidden assumptions about the audience, brand tone, device context, and whether the user is asking for brand exploration or product-flow exploration.
4. If the request is under-specified, define 2-3 plausible directions and recommend one, making clear how each direction changes the artifact type or user flow.
5. Identify what must stay stable: flows, content hierarchy, design system boundaries, accessibility constraints, or upstream-merge boundaries.
6. Choose the visual direction and express it in concrete terms:
   - typography
   - color
   - layout
   - motion
   - background treatment
7. State the deliverable type explicitly before implementation or handoff: visual mockup, clickable prototype, or implementation-ready spec.
8. Implement or guide implementation in a way that keeps the visual language consistent across the touched surface.
9. Verify the result in the browser when possible, especially for spacing, hierarchy, responsive behavior, and interaction polish.

## Output

Report:

- UI problem being solved
- artifact type chosen
- visual direction chosen
- key aesthetic decisions
- real pages or user flows the design is covering
- stability constraints preserved
- whether upstream-merge boundaries or reference-reuse boundaries influenced the design
- browser-visible evidence or verification path
- remaining risks, especially where the UI may still feel generic

From gclm/super-stack, MIT.

More from gclm/super-stack

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Mar 29, 2026 (7 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from gclm

Cite this page

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