agentleFS
MADTeacherMADTeacher/mad-agents-skills/flutter-adaptive-ui/SKILL.md

flutter-adaptive-ui

Skill106 starsChanged 7 months agoMITView on GitHub

Build, fix, review, and validate adaptive or responsive Flutter UIs for mobile, tablet, desktop, web, large screens, foldables, and mixed input devices. Use when creating breakpoint-driven layouts, responsive navigation, adaptive dialogs/lists/grids, keyboard/mouse/touch behavior, window-size decisions with MediaQuery or LayoutBuilder, or Capability and Policy patterns for platform-specific behavior.

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 8 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    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. Very specific:

    Use MediaQuery.sizeOf(context) for app-level or page-level window decisions.

    line 31
  2. Very specific:

    Use LayoutBuilder when the branch depends on the parent constraints of a widget subtree.

    line 32
  3. Distinctive:

    Do not read every reference by default.

    line 50
  4. Very specific:

    Treat Dart snippets in references/ as explanatory fragments unless the reference explicitly says otherwise.

    line 54
  5. Very specific:

    Do not use Platform.isIOS, Platform.isAndroid, device names, or OrientationBuilder for layout decisions.

    line 55
  6. Distinctive:

    Do not let large screens stretch text fields, cards, lists, or reading content across the full window without a deliberate max width or …

    line 56
  7. Distinctive:

    Start with a solid touch interaction, then add hover, shortcuts, focus traversal, and keyboard activation as accelerators.

    line 57
  8. Very specific:

    Use GridView.extent, adaptive flex layouts, or constrained content widths for expanded layouts instead of duplicating whole screens when a …

    line 58
  9. Distinctive:

    Keep Capability methods about what is possible and Policy methods about what should be shown or allowed.

    line 59
  10. Very specific:

    Treat bundled assets as starter examples.

    line 60
  11. Distinctive:

    After changing a Flutter project:

    line 64
  12. Distinctive:

    Run relevant widget tests when layout branching, navigation state, focus, or policy/capability behavior changed.

    line 67
  13. Distinctive:

    Verify no new overflow stripes, clipped text, lost selected state, lost scroll position, or broken keyboard traversal.

    line 69
  14. Distinctive:

    If validation cannot run, report the blocker and the risk.

    line 70

≈ “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

Flutter Adaptive UI

You are a Flutter adaptive UI implementation agent. Your job is to make the UI work from narrow mobile windows to expanded desktop/web layouts without guessing from device type.

Principle 0

Adaptive Flutter UI is based on available constraints, not platform labels. Use window or parent constraints for layout decisions, keep touch usable first, and add mouse, keyboard, and platform behavior as explicit branches that can be tested.

Core rule: constraints go down, sizes go up, parent sets position.

Default breakpoints:

  • Compact: width < 600
  • Medium: 600 <= width < 840
  • Expanded: width >= 840

Workflow

  1. Identify the user flow, target form factors, supported platforms, and the expensive failure mode: overflow, unreadable wide content, lost state, inaccessible keyboard flow, or wrong platform behavior.
  2. Inspect existing widgets before changing layout. Find navigation, dialogs, lists, grids, fixed widths/heights, orientation checks, Platform.* layout checks, and custom input/focus handling.
  3. Abstract shared data before branching. For example, create one destination model used by both NavigationBar and NavigationRail.
  4. Measure the right space:
    • Use MediaQuery.sizeOf(context) for app-level or page-level window decisions.
    • Use LayoutBuilder when the branch depends on the parent constraints of a widget subtree.
  5. Branch by breakpoints or capabilities, not by device type. Use compact/medium/expanded layouts for space changes; use Capability and Policy objects for what the app can do or should do.
  6. Implement the smallest adaptive change that preserves state. Keep scroll position, selected navigation destination, form input, and focus stable across resize/orientation/fold changes.
  7. Validate on at least compact and expanded widths. Include medium width when the implementation has a distinct tablet layout.

Resource Routing

TaskRead/usePurpose
Need the full adaptive design processadaptive-workflow.mdAbstract, measure, branch workflow and breakpoint choice
Need constraints or overflow diagnosislayout-constraints.mdFlutter layout rules and edge cases
Need basic layout widget guidancelayout-basics.mdRows, columns, alignment, sizing, and composition
Need common layout widget behaviorlayout-common-widgets.mdContainer, GridView, ListView, Stack, Card, ListTile
Need adaptive UX guardrailsadaptive-best-practices.mdOrientation, width, inputs, state, and performance guidance
Need platform behavior branchingadaptive-capabilities.mdCapability and Policy structure
Need responsive navigation starter coderesponsive_navigation.dartCopy only after fitting destinations and selected state to the target app
Need Capability/Policy starter codecapability_policy_example.dartCopy only after replacing placeholder behavior with real app services

Do not read every reference by default. Read only the routed material needed for the current failure mode or implementation path.

Implementation Rules

  • Treat Dart snippets in references/ as explanatory fragments unless the reference explicitly says otherwise. Use assets/ for starter code.
  • Do not use Platform.isIOS, Platform.isAndroid, device names, or OrientationBuilder for layout decisions. Use available width/constraints instead.
  • Do not let large screens stretch text fields, cards, lists, or reading content across the full window without a deliberate max width or multi-column layout.
  • Start with a solid touch interaction, then add hover, shortcuts, focus traversal, and keyboard activation as accelerators.
  • Use GridView.extent, adaptive flex layouts, or constrained content widths for expanded layouts instead of duplicating whole screens when a local reflow is enough.
  • Keep Capability methods about what is possible and Policy methods about what should be shown or allowed. Name methods by the decision, not by the platform.
  • Treat bundled assets as starter examples. They are copy-ready only after flutter analyze passes in the target project or a scratch Flutter project.

Validation

After changing a Flutter project:

  1. Run flutter analyze.
  2. Run relevant widget tests when layout branching, navigation state, focus, or policy/capability behavior changed.
  3. Manually or automatically check narrow (<600), medium (600-839 when used), and expanded (>=840) widths.
  4. Verify no new overflow stripes, clipped text, lost selected state, lost scroll position, or broken keyboard traversal.
  5. If validation cannot run, report the blocker and the risk. Do not present an adaptive UI change as verified without a width check or analysis result.

Fallback

If the target project lacks enough context to choose a final layout, implement the smallest reversible abstraction first: shared destination/data models, local LayoutBuilder branches, and isolated Capability/Policy interfaces. Ask the user only for product decisions that cannot be inferred from the app, such as which content should move, hide, or become primary on each form factor.

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.