agentleFS
KyaniteLabsKyaniteLabs/tastecheck/skills/responsive-layout/SKILL.md

responsive-layout

Skill12 starsChanged 11 days agoMITView on GitHub

Use when a page or component must survive narrow containers, long or translated content, zoom, reflow, or overflow without device-name breakpoints.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 5 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

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

    Start mobile-first with Grid/Flex intrinsic sizing (minmax, auto-fit, clamp, %/fr).

    line 15
  2. Distinctive:

    Add breakpoints only where content breaks.

    line 16
  3. Distinctive:

    Verify minimum supported viewport, 200–400% zoom, long/translated content, narrow embeds, and wide views without overflow or collapse.

    line 17
  4. Very specific:

    Start with non-disappearing content, reading order, and narrowest real container.

    line 22
  5. Distinctive:

    Start from the narrowest supported content state.

    line 28
  6. Distinctive:

    Prefer container queries for reusable components when their behavior depends on their container, not the viewport.

    line 30
  7. Very specific:

    Use references/patterns.md for grid, sidebar, stack, cluster, and container-query patterns.

    line 38
  8. Very specific:

    Start with minmax(min(16rem, 100%), 1fr), min-width: 0, logical sizing, and content-led breakpoints.

    line 39
  9. Distinctive:

    Use five-field ledger; start Reason with check ID.

    line 50
  10. Very specific:

    Stop / handoff: Pause when content priority is unknown and reflow would silently hide required information (+1 in contract.json).

    line 78

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Responsive Layout

Responsive layout is intrinsic reflow plus content-led breakpoints/container queries, not named-device CSS.

The decision order

  1. Start mobile-first with Grid/Flex intrinsic sizing (minmax, auto-fit, clamp, %/fr).
  2. Add breakpoints only where content breaks; reusable components use container queries.
  3. Verify minimum supported viewport, 200–400% zoom, long/translated content, narrow embeds, and wide views without overflow or collapse.

Preserve hierarchy under pressure

Start with non-disappearing content, reading order, and narrowest real container. Choose stack/reorder/progressive reveal/rail from observed pressure; test long content, locale, zoom, and embeds before calling it responsive.

Non-negotiables

  • Start from the narrowest supported content state; avoid device-name breakpoints.
  • Prevent overflow with min-width: 0, bounded media, and wrapping; reflow rather than shrink to unusable.
  • Prefer container queries for reusable components when their behavior depends on their container, not the viewport.
  • At narrow widths and zoom, reading/interaction should not require two-dimensional scrolling. Isolate essential exceptions (wide tables, maps, timelines, canvases) in a labelled, keyboard-operable scroll region.

Quick-start patterns

Use references/patterns.md for grid, sidebar, stack, cluster, and container-query patterns. Start with minmax(min(16rem, 100%), 1fr), min-width: 0, logical sizing, and content-led breakpoints.

Reference files

  • assets/layout-patterns.css — intrinsic examples after pressure evidence.
  • references/patterns.md — patterns, units, images, and overflow debugging.
  • references/decision-records.md — novel-case ADR rules.

Evidence

Use five-field ledger; start Reason with check ID.

StatusReasonRemediationEvidenceProvenance
layout-strategy — content hierarchy and intrinsic composition
containment — breakpoint and container-query rationale
narrow-reflow — 320px, long-content, and overflow result
zoom-interaction — zoom and keyboard-reachable interaction result
handoff — component and accessibility boundary

Pressure test

Each row names longest-pressure fixture, width/zoom, expected and observed result (or fail-closed missing evidence). Test full shell, narrow, 400%, and narrow embed; record what stays first/stacks/folds and why. No rendered build means plan, not pass.

How to deliver

Deliver content-led layout/breakpoints and rerunnable narrow/long/container/zoom/keyboard evidence; hand type, rhythm, and state concerns onward.

<!-- contract:v1:start -->

Contract (generated)

Canonical detail: contract.json.

  • Route: A layout needs intrinsic reflow, responsive hierarchy, container behavior, or overflow repair (+1 in contract.json); avoid: The request is only to choose a spacing scale or type system (+1 in contract.json)
  • Exclude: Do not choose breakpoints from device names (+1 in contract.json)
  • Stop / handoff: Pause when content priority is unknown and reflow would silently hide required information (+1 in contract.json); receives [art-direction, design-system-interview, improve-existing-website, spacing-system, tasteroll, theming, web-typography] -> sends [component-states, form-ux, a11y-pass, deslop-ui, i18n-ready]
  • Output: An intrinsic layout strategy with justified breakpoints and responsive verification evidence
  • Evidence: table_with_evidence with status, reason, remediation, evidence, provenance.
<!-- contract:v1:end -->

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.