agentleFS
LiriRaidLiriRaid/waymark/skills/ui-system/SKILL.md

ui-system

Skill2 starsChanged 6 days agoMITView on GitHub

owner dept-ux-ui. Design systems and identity: \"paleta de colores\", \"tipografía\", \"design tokens\", \"tema oscuro\", \"sistema de diseño\", \"qué estilo le queda\".

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Memory is the system's spine: Every decision lands in project memory (## Design system) so the next session extends it.

    line 16
  2. Very specific:

    Before building, list patterns/ and read the matching file, if any.

    line 32
  3. Very specific:

    Update an existing pattern only with new, verified information (novelty check, waymark references/learning.md).

    line 34
  4. Distinctive:

    Read only the mode the task needs (one file).

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

UI System

Precondition. Tool of dept-ux-ui. If that department skill is not loaded in this conversation, load it first and use its brief (what, why, where, how) as the input of this skill. Skip only for L0 edits.

Approach

A design system is derived from a few decisions, not chosen from a catalogue: product, audience, conditions of use and existing code in; tokens in the project's own styling system out.

  • Existing system first. Discover and record before proposing. Ignoring what shipped creates a second system.
  • Few inputs, many outputs. One brand hue, one scale ratio and one base unit generate palette, type and spacing.
  • Roles over values. Components consume semantic roles (surface, text-muted, danger), never raw values or primitive steps.
  • Accessibility constrains generation. A color pair becomes a token only after its contrast passes in every theme it appears in.
  • Memory is the system's spine. Every decision lands in project memory (## Design system) so the next session extends it.
  • Questions before acting: Is there a system already, and where is its source of truth? Who uses it, how long, on which devices? Light, dark or both? Which component library constrains tokens? What brand traits must stay?

Inputs

InputSource
Briefthe dept-ux-ui brief
Stack conventions../waymark/stacks/<stack>.md → Frontend (styling system, component library, SSR)
Project context<project>/.waymark/memory.md → ## Design system (this skill's record, shared with ui-build and ui-refine)
Methodsreferences/: palette-method, type-scale, product-direction, chart-choice (load only what the mode needs)
Known patternspatterns/ in this skill (generic) + project skills (project-specific)

Output contract

Always return: mode(s) run; sources of truth read; files changed; decisions (recorded in project memory: yes/no); contrast table for every touched pair (pair · theme · ratio · pass/fail); drift counts; open risks (failing pairs, unmigrated files, library tokens not covered); hand-offs (dept-frontend for runtime theme code, ui-audit for a full accessibility pass, dept-qa for visual regression).

Pattern library (grows with use)

  • Before building, list patterns/ and read the matching file, if any. See patterns/README.md for what belongs here.
  • After building something reusable that has no pattern yet (theme switch, status color set, KPI tile row…), write patterns/<pattern>.md from ../waymark/templates/pattern.template.md: stack-agnostic intent, anatomy, states, a11y, pitfalls, and one short adapter per stack it was built in. Project-specific details go to project memory, not here.
  • Update an existing pattern only with new, verified information (novelty check, waymark references/learning.md).

Modes

Read only the mode the task needs (one file); never load all modes.

  • discover — first UI-system task in a project; record missing or older than the theme files; always before recommend or tokens on existing code. → modes/discover.md
  • recommend — greenfield, explicit redesign, "qué estilo le queda", or "hazlo más profesional" when the system itself is the problem rather than one screen (one scr → modes/recommend.md
  • tokens — implement an approved system, add a theme ("tema oscuro"), or refactor tokens ("unifica los colores", "design tokens"). → modes/tokens.md
  • chart — "qué gráfica uso", dashboards, KPI rows, choosing chart colors. → modes/chart.md
  • consistency-check — before a tokens refactor, after a large UI change, or "se ve inconsistente". Read-only unless the user asks to fix (then tokens). → modes/consistency-check.md

Stack adapters

Stack-specific notes → references/stack-adapters.md (read only the project's stack row).

Learned notes

Grows with use (waymark references/learning.md). Dated, non-obvious notes about using this tool. When there are more than ~10, fold them into the body above and clear this list.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.