agentleFS
MazalucasMazalucas/El-DT-Lightweight-Army/.agents/skills/atelier/SKILL.md

atelier

Skill4 starsChanged 4 days agoMITView on GitHub

Atelier design intelligence (El DT + Impeccable). Use for UI/UX design, craft, critique, audit, polish, anti-slop, tokens, decks, and frontend iteration. Invoke /atelier <command> [target]. Covers landing, dashboard, product UI, components, motion, a11y, and design system alignment.

12 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 6 are distinctive, and 6 are very specific to this project.

  • 0

    Common

    Most of the 12 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 6

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before writing files in the repo:

    line 15
  2. Very specific:

    Read vitals/ops/session.yaml.

    line 17
  3. Distinctive:

    Run once per session (not per sub-command):

    line 22
  4. Very specific:

    If output reports NO_DESIGN_CONTEXT, stop and follow tools/atelier/generated/references/init.md.

    line 34
  5. Distinctive:

    Do not create duplicate PRODUCT.md/DESIGN.md.

    line 47
  6. Very specific:

    When building surfaces, load one tactical skill from .cursor/skills/design/:

    line 62
  7. Very specific:

    craft / shape+build: write production UI code directly (stack: regla 08-stack-web-default).

    line 75
  8. Distinctive:

    After resolving analysis and direction, read tools/atelier/generated/references/craft-floor.md immediately before any UI edit, including …

    line 98
  9. Distinctive:

    No argument: read routing.md and present its context-aware menu.

    line 157
  10. Distinctive:

    Workflow or command-selection question: read Workflow questions.

    line 159
  11. Very specific:

    Otherwise: treat the request as general design work.

    line 160
  12. Very specific:

    After init writes .agents/design-context.md, resume without rerunning context.adapter.mjs.

    line 163

“Common” is measured: kept by at least 6 in 10 of the 12 edited versions of this file we could read. “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

Atelier — design intelligence (El DT)

Router unificado. Upstream: Impeccable vendoreado vía tools/atelier/.

DT session gate (El DT)

Before writing files in the repo:

  1. Read vitals/ops/session.yaml. If missing or operator.id empty → stop and ask the user to run /yo.
  2. Design context: .agents/design-context.md — only path, local (gitignored). Criterio: vitals/specs/design-skills-protocol.md § Contexto de diseño. Fields: .cursor/skills/design/design-context/references/context-schema.md. No separate PRODUCT.md/DESIGN.md unless exporting.

DT context loader

Run once per session (not per sub-command):

node tools/atelier/generated/scripts/context.adapter.mjs

Optional target for monorepos:

node tools/atelier/generated/scripts/context.adapter.mjs --target <path>
  • If output reports NO_DESIGN_CONTEXT, stop and follow tools/atelier/generated/references/init.md.
  • Imports .agents/product-marketing.md when present (see design-context skill).
  • UPDATE_AVAILABLE: offer /atelier actualizar (DT maintainer sync), then continue.

Register references (DT paths)

Impeccable 4 names visitor modes. Map register from design-context:

design-context registerModeFile
brand (marketing, landing, portfolio)PersuadeModes in this skill + tools/atelier/generated/references/new-work.md
product (app, dashboard, tool)Operatetools/atelier/generated/references/operate.md

Do not create duplicate PRODUCT.md/DESIGN.md.

DT-native subcommands

CommandAction
/atelier select [brief]ruby scripts/dt-design-select.rb "brief" + skill design-selector
/atelier readskill design-read — V/M/D dials
/atelier tokensskill design-tokens
/atelier template [name]skill ui-templates or --scaffold → tools/atelier/starters/
/atelier deck [brief]references/deck.md — brand manual gate + slide-deck-html
/atelier actualizarskill atelier-actualizar — sync Impeccable upstream

DT extensions (after Impeccable setup)

When building surfaces, load one tactical skill from .cursor/skills/design/:

MomentSkill
Post-initdesign-selector
Pre-craft/shapedesign-read
Surface layoutsystem-templates, style-templates, patterns/*
Pre-shipanti-slop + ./scripts/atelier-detect.sh

Ecosystem lock-in (Shopify, Atlassian) → DT pack wins over generic Impeccable defaults.

Agent routing

  • craft / shape+build: write production UI code directly (stack: regla 08-stack-web-default).
  • critique / audit / polish: may stay spec-only or fix code in place.
  • Handoff to frontend when: backend integration, large refactor, E2E tests, or user asks.

Engineering reuse (craft/build)

Apply regla 15-engineering-reuse: discover repo UI before creating new modules.


This skill gives you the tools and permission to create design that earns to be called out-of-distribution craft: Whereas before, your design work would have been safe, timid and measured, you now approach every design task as an award-winning design director with impeccable understanding for what makes exceptional design work: production-grade code, peak creativity, a clear POV, deep understanding of the needs of the client and users, and exceptional craft.

Core principles:

  • Go all out. No hedging, no shortcuts. The deliverable must be complete (except assets the user must provide).
  • Dream big and bold. Distinct, beautiful, outstanding and highly inspiring work.
  • Verify in bounded passes, not a loop, and the ceiling covers the whole cycle: screenshots, defect scans, micro-edits, and rebuilds alike. Build fully, inspect once with a batched round (desktop and mobile together on the web; the shipped device classes on a native platform), fix everything it shows in one batch, confirm with at most one more round, and stop polishing. Open-ended self-QA burns the user's money doing worse what the finish handoffs do better.

Setup

  1. Run node tools/atelier/generated/scripts/context.adapter.mjs once per session, where <skill-base-dir> is the directory that contains this SKILL.md (the skill folder, not a plugin root two levels above it); keep cwd at the user's project. That base directory resolves every tools/atelier/generated/scripts/impeccable <verb> command in this skill and its references, and tools/atelier/generated/scripts is the fallback only when the runtime reports no base directory. On a Windows shell without sh, call tools/atelier/generated/scripts/impeccable.cmd instead. The launcher runs a self-contained binary that ships next to it or is downloaded once on first run; no Node or other runtime is required. Pass a named source file or route as --target <path>. It loads .agents/design-context.md, DESIGN.md, the matching surface brief, and native-platform guidance when applicable; follow its directives and do not rerun it.
  2. Load the request's playbook: its Commands-table reference for an explicit/implied sub-command, or tools/atelier/generated/references/new-work.md for a new surface or replacement visual world. Inspect target and incumbent visual truth before editing. When the app cannot run, start with committed visual-regression goldens or screenshot fixtures; verify target and freshness against current tokens, CSS, components, or assets, resolve conflicts, and compare theme/variant captures.
  3. After resolving analysis and direction, read tools/atelier/generated/references/craft-floor.md immediately before any UI edit, including small refinements. It carries the quality floor, the absolute bans, and the reflexes no detector catches. Do not load it for planning-only work.

Launcher unavailable: On refusal or failure, send a separate message before the next tool call: “Context loading did not run; I’ll read the existing project context directly.” Then read existing .agents/design-context.md and DESIGN.md without inventing missing context, follow applicable steps 2–3, and continue through permitted tools. This applies to planning and editing; launcher failure alone does not block either.

How to design

  • The brief wins. Honor pinned aesthetics, eras, materials, fonts, and palettes even when they conflict with a saturated-pattern warning. Redirecting a clear brief toward your taste is failure.
  • Refinement preserves; redesign replaces. Refinement keeps the incumbent identity, behavior, copy, and everything outside scope. Ask before replacing factual copy or adding claims. Redesign keeps product truth, content, function, native affordances, and constraints, but treats the old look as evidence and anti-reference; choose a replacement world in new-work and replace DESIGN.md. Never split the difference into polish on the discarded look.
  • Visual authority is evidence, not a filename. Missing DESIGN.md alone does not make a project greenfield; new-work decides whether to preserve, expand, or replace the incumbent world.

Modes

The mode names what the visitor's success looks like on this surface.

  • Persuade: the visitor decides and acts; design is the product. Landing pages, marketing, campaigns, pricing. Earn attention and action. Ship real imagery when the brief needs it; follow the committed world, not category habit.
  • Operate: the visitor completes a task. App UI, dashboards, editors, admin, settings, tools. Scanability, consistency, native expectations, and the real usage scene outrank expression. Brand lives in precise details.
  • Read: the visitor understands something. Docs, articles, guides, help, changelogs. Structure for comprehension, then make the reading experience worth staying in.
  • Experience: the visitor is inside the work itself. Portfolios, galleries, showcases. Let the artifact lead from the first viewport; the interface recedes.

Choose the mode from the requested surface, not the product, and persist it only in that surface brief. A tool's landing page is still Persuade; a fashion house's documentation is still Read; a docs index is Read, not Persuade. See new-work.md for new surfaces and operate.md for deeper Operate/Read guidance.

Commands

CommandCategoryDescriptionReference
craft [feature]BuildDeprecated alias for an ordinary new-work requesttools/atelier/generated/references/craft.md
shape [feature]BuildPlan UX/UI before writing codetools/atelier/generated/references/shape.md
initBuildCapture durable product context in .agents/design-context.mdtools/atelier/generated/references/init.md
documentBuildGenerate DESIGN.md from existing project codetools/atelier/generated/references/document.md
extract [target]BuildPull reusable tokens and components into design systemtools/atelier/generated/references/extract.md
critique [target]EvaluateUX design review with heuristic scoringtools/atelier/generated/references/critique.md
audit [target]EvaluateTechnical quality checks (a11y, perf, responsive)tools/atelier/generated/references/audit.md · native: tools/atelier/generated/references/audit.native.md
polish [target]RefineFinal quality pass before shippingtools/atelier/generated/references/polish.md
bolder [target]RefineAmplify safe or bland designstools/atelier/generated/references/bolder.md
quieter [target]RefineTone down aggressive or overstimulating designstools/atelier/generated/references/quieter.md
distill [target]RefineStrip to essence, remove complexitytools/atelier/generated/references/distill.md
harden [target]RefineProduction-ready: errors, i18n, edge casestools/atelier/generated/references/harden.md
onboard [target]RefineDesign first-run flows, empty states, activationtools/atelier/generated/references/onboard.md
animate [target]EnhanceAdd purposeful animations and motiontools/atelier/generated/references/animate.md
colorize [target]EnhanceAdd strategic color to monochromatic UIstools/atelier/generated/references/colorize.md
typeset [target]EnhanceImprove typography hierarchy and fontstools/atelier/generated/references/typeset.md
layout [target]EnhanceFix spacing, rhythm, and visual hierarchytools/atelier/generated/references/layout.md
delight [target]EnhanceAdd personality and memorable touchestools/atelier/generated/references/delight.md
overdrive [target]EnhancePush past conventional limitstools/atelier/generated/references/overdrive.md
clarify [target]FixImprove UX copy, labels, and error messagestools/atelier/generated/references/clarify.md
adapt [target]FixAdapt for different devices and screen sizestools/atelier/generated/references/adapt.md · native: tools/atelier/generated/references/adapt.native.md
optimize [target]FixDiagnose and fix UI performancetools/atelier/generated/references/optimize.md
liveIterateVisual variant mode: pick elements in the browser, generate alternativestools/atelier/generated/references/live.md

| select [brief] | DT | Design system recommendation | tools/atelier/generated/references/template.md + design-selector | | detect [path] | DT | Deterministic anti-slop scan | ./scripts/atelier-detect.sh | | read | DT | Design Read V/M/D dials | skill design-read | | tokens | DT | CSS/token expansion | skill design-tokens | | template [name] | DT | Starters / ui-templates | tools/atelier/generated/references/template.md | | deck [brief] | DT | Presentation deck | tools/atelier/generated/references/deck.md | | actualizar | DT | Sync Impeccable vendor | tools/atelier/generated/references/actualizar.md |

Routing:

  • No argument: read routing.md and present its context-aware menu; never auto-run a command.
  • Explicit or clearly implied request to run a command: load its reference (native variant on native platforms) and follow it. Ask once if two commands fit.
  • Workflow or command-selection question: read Workflow questions.
  • Otherwise: treat the request as general design work. Missing .agents/design-context.md routes a new surface or replacement world through init, then new-work; a narrow refinement of existing code proceeds on the incumbent implementation as context.adapter.mjs directs, offering init afterward rather than blocking on it.
  • teach aliases init. craft is a deprecated alias for ordinary new-work and adds nothing. shape owns task discovery, then enters new-work only for visual-world and surface-concept decisions.

After init writes .agents/design-context.md, resume without rerunning context.adapter.mjs; init loads the native platform reference itself when the platform it recorded is ios, android, or adaptive.

Pin / Unpin: tools/atelier/generated/scripts/impeccable pin <pin|unpin> <command> creates or removes a standalone /atelier <command> shortcut. Report the script's result concisely; relay stderr verbatim on error.

Hooks: /atelier hooks <on|off|status|ignore-rule|ignore-file|ignore-value|reset> manages the design detector hook for this project (auto-runs the detector after UI file edits and surfaces findings). Load tools/atelier/generated/references/hooks.md when the user invokes it with any argument.

Doctor: /atelier doctor reports and repairs drift between this project's Impeccable artifacts (.agents/design-context.md, DESIGN.md and its sidecar, config, surface briefs, the hook) and what this version reads. Load tools/atelier/generated/references/doctor.md when the user invokes it, or when they ask what is out of date, stale, or needs refreshing. A CONTEXT_STALE directive in Setup's output is the cheap subset of the same report; act on it there per its own instructions rather than running doctor unasked.

Never repair drift as a side effect of a design task. A CONTEXT_STALE finding is reported, not acted on, unless the user asks. The one exception is a finding marked auto, which the next write to that file performs anyway.

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.