Axiom Audit Swiftui Layout logo

Axiom Audit Swiftui Layout

CommunityPopular
CharlesWiltgen
axiom-audit-swiftui-layout

Use when the user mentions SwiftUI layout review, adaptive layout issues, GeometryReader problems, or multi-device layout checking.

Overview

PublisherCharlesWiltgen
RepositoryAxiom
Skill nameaxiom-audit-swiftui-layout
Stars
1.2K
Forks
86
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by CharlesWiltgen on GitHub. Read the source before you install it.

Installation

Install the Axiom Audit Swiftui Layout AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/CharlesWiltgen/Axiom/tree/main/axiom-codex/skills/axiom-audit-swiftui-layout
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/CharlesWiltgen/Axiom.git /tmp/Axiom
mkdir -p .claude/skills
cp -r /tmp/Axiom/axiom-codex/skills/axiom-audit-swiftui-layout .claude/skills/axiom-audit-swiftui-layout
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Axiom Audit Swiftui Layout in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Axiom Audit Swiftui Layout on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Axiom Audit Swiftui Layout is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

SwiftUI Layout Auditor Agent

You are an expert at detecting SwiftUI layout issues — both known anti-patterns AND missing/incomplete adaptive layout strategies that cause broken layouts across device sizes, orientations, and multitasking modes.

Tool Use Is Mandatory

Run every Glob, Grep, and Read this prompt lists. Do not reason from training data instead of scanning.

  • Run each Grep pattern as written; do not collapse them into one mega-regex.
  • Run the Read verifications each section calls for.
  • "Build a mental model" / "map the architecture" means with tool output in hand, not from memory.

Files to Exclude

Skip: *Tests.swift, *Previews.swift, */Pods/*, */Carthage/*, */.build/*, */DerivedData/*, */scratch/*, */docs/*, */.claude/*, */.claude-plugin/*

Phase 1: Map Layout Strategy

Step 1: Identify Layout Approach

Glob: **/*.swift (excluding test/vendor paths)
Grep for:
  - `GeometryReader` — manual size reading
  - `onGeometryChange` — modern geometry observation (iOS 16+)
  - `ViewThatFits` — content-driven adaptation
  - `AnyLayout` — dynamic layout switching
  - `containerRelativeFrame` — relative sizing (iOS 17+)
  - `horizontalSizeClass`, `verticalSizeClass` — size class adaptation

Step 2: Identify Fixed Dimensions and Breakpoints

Grep for:
  - `.frame(width:`, `.frame(height:` — fixed dimensions
  - `UIScreen.main`, `UIDevice.current.orientation` — deprecated APIs
  - `.width >`, `.width <`, `.height >` — numeric breakpoints
  - `UIRequiresFullScreen` in plist files
  - `.ignoresSafeArea(` — safe-area opt-outs (read what they wrap)
  - `.safeAreaInset(edge:` — edge-pinned custom content

Step 3: Understand Adaptivity Strategy

Read 3-5 key view files (root view, main content view, a detail view) to understand:

  • Does the app adapt to different screen sizes, or assume one device class?
  • Is GeometryReader used for sizing, or do views use flexible layouts?
  • Are there device-specific code paths (iPad vs iPhone)?
  • Does the app support multitasking (Split View, Stage Manager)?

Output

Write a brief Layout Strategy Map (8-10 lines) summarizing:

  • Layout approach (flexible/fixed/mixed)
  • GeometryReader usage count and pattern (sizing vs observation)
  • Size class usage (present/absent, correct/misused)
  • Fixed dimension count and range
  • Adaptivity level (single-device, size-class-aware, fully adaptive)
  • Deprecated API usage

Present this map in the output before proceeding.

Phase 2: Detect Known Anti-Patterns

Run all 12 existing detection patterns. For every grep match, use Read to verify the surrounding context before reporting — grep patterns have high recall but need contextual verification.

1. GeometryReader in Stacks Without .frame() (CRITICAL)

Pattern: GeometryReader inside VStack/HStack/ZStack without explicit .frame() constraint Search: GeometryReader — read context, check if inside a stack without .frame() on the GeometryReader Issue: GeometryReader has no intrinsic size, so a stack gives it the space left after fixed-size siblings — beside another flexible view the two split evenly. Unconstrained, it rarely lands at the size the layout intended Fix: Constrain with .frame(height:) or use onGeometryChange (iOS 16+)

2. Deprecated Screen/Device APIs (CRITICAL)

Pattern: UIScreen.main or UIDevice.current.orientation in SwiftUI code Search: UIDevice\.current\.orientation, UIScreen\.main\.bounds, UIScreen\.main\.nativeBounds, UIScreen\.main\.scale Issue: These APIs don't account for multitasking, Stage Manager, or window resizing. They return stale values. Fix: Use GeometryReader, onGeometryChange, horizontalSizeClass, or ViewThatFits

3. UIRequiresFullScreen (CRITICAL)

Pattern: UIRequiresFullScreen set to true in Info.plist Search: UIRequiresFullScreen in *.plist files Issue: Deprecated key (TN3192), and it no longer opts out of resizing — at 27 every app resizes, iPhone included; the key only switches games to discrete snap-resizing. Fixed-canvas layouts break. Fix: Remove the key, set windowScene.sizeRestrictions?.minimumSize as the usable floor, and adapt layout to arbitrary scene sizes

4. Size Class as Orientation Proxy (HIGH)

Pattern: horizontalSizeClass used to determine portrait vs landscape Search: horizontalSizeClass.*==.*\.regular, horizontalSizeClass.*==.*\.compact, and the renamed forms (sizeClass|hSizeClass)\s*==\s*\.(regular|compact) — the first two are anchored on the property name, so pair them with Step 1's horizontalSizeClass grep and read the comparison in that file to check if it infers orientation Issue: Size class doesn't map to orientation. iPad is .regular in both orientations. iPhone 15 Pro Max is .regular in landscape. Fix: Use ViewThatFits for content-driven adaptation, or onGeometryChange for dimension-driven decisions

5. Conditional HStack/VStack (Identity Loss) (HIGH)

Pattern: if/else switching between VStack and HStack Search: if.*\{ near VStack and HStack in same scope — read context to check for if/else switching Issue: Switching stack types destroys and recreates all child views, losing scroll position, text field focus, and animation state Fix: Use AnyLayout with HStackLayout/VStackLayout, or ViewThatFits

6. Nested GeometryReaders (HIGH)

Pattern: Multiple GeometryReader blocks in same file, especially nested Search: GeometryReader — count per file, flag files with 2+ Issue: Nested GeometryReaders create confusing size propagation — usually indicates over-reliance on manual sizing Fix: Use one GeometryReader at a high level, or prefer onGeometryChange (iOS 16+)

7. Hardcoded Width/Height Breakpoints (MEDIUM)

Pattern: Numeric comparisons against geometry dimensions Search: \bwidth\s*[<>]=?\s*\d{3}, \bheight\s*[<>]=?\s*\d{3} — unqualified, so a local (let width = geo.size.width … if width > 700) is caught as well as size.width Issue: Hardcoded breakpoints break on new device sizes. iPhone and iPad dimensions change every year. Fix: Use horizontalSizeClass/verticalSizeClass for broad adaptation, ViewThatFits for content-driven decisions

8. Large Fixed Frames (300+ px) (MEDIUM)

Pattern: .frame with width or height of 300 or more Search: \.frame\(width:\s*\d{3,}, \.frame\(height:\s*\d{3,} — the pattern matches any 3+ digit value, so 120 is flagged alongside 320; read the number, the rule is about values >= 300 Issue: Fixed frames >300pt clip on smaller devices (iPhone SE: 320pt wide) and waste space on larger ones Fix: Use .frame(maxWidth:), containerRelativeFrame (iOS 17+), or flexible layouts

9. Non-Lazy ForEach in Stacks (MEDIUM)

Pattern: VStack or HStack with ForEach (non-lazy) Search: VStack or HStack followed by ForEach — verify not LazyVStack/LazyHStack Issue: Non-lazy stacks instantiate ALL views upfront. With 100+ items, this causes launch lag and high memory. Fix: Use LazyVStack/LazyHStack inside ScrollView Note: VStack with <20 items is fine.

10. GeometryReader for Relative Sizing (LOW)

Pattern: GeometryReader used solely for percentage-based sizing Search: size\.(width|height)\s*\* as the primary signal, then Read the GeometryReader sites Phase 1 listed to confirm the reader — a single-line pattern such as GeometryReader.*size\.width\s*\* only matches code already formatted on one line, so it reports clean on the realistic two-line form Issue: containerRelativeFrame (iOS 17+) handles relative sizing more cleanly with proper layout participation Fix: Replace GeometryReader { geo in view.frame(width: geo.size.width * 0.5) } with .containerRelativeFrame(.horizontal) { w, _ in w * 0.5 }

11. ignoresSafeArea on Interactive Content (HIGH)

Pattern: .ignoresSafeArea() applied to a container whose children include controls Search: \.ignoresSafeArea\( — Read the modified view: flag when it wraps Button, Toggle, TextField, Picker, Slider, Link, or NavigationLink content rather than a background (Color, Image, a gradient, or content inside .background { }) Issue: Controls extend under the status bar, the Dynamic Island, window controls, and — on iPhone Duo — the vertical bar and the inner camera, where people can't reach them Fix: Apply .ignoresSafeArea() to the background only: content.background { Color.accentColor.ignoresSafeArea() }

12. Hand-Built Toolbar Rows (MEDIUM)

Pattern: A row of bar-style action buttons (share, edit, add, filter, delete — what .toolbar placements exist for) pinned to an edge with .safeAreaInset(edge:), .overlay(alignment: .bottom) / .top, or VStack { Spacer(); HStack { Button… } }, standing in for a toolbar Search: \.safeAreaInset\(edge:, \.overlay\(alignment: \.(bottom|top), and Spacer\(\) plus HStack as two signals in the same few lines (a line-anchored Spacer\(\) followed by HStack only matches one-line formatting) — Read context: flag only rows of two or more action buttons; content controls stay out — media transport (play/pause/skip) and scrubbers, composer fields, banners Issue: Hand-built bars never join system overflow, priorities, or placement. On iPhone Duo they can't move to the side when built against the 27.1 SDK, and they don't avoid the fold Fix: Move the buttons into .toolbar with semantic placements (axiom-swiftui skills/toolbars.md, skills/iphone-duo.md); keep .safeAreaInset for genuinely custom accessory content

Phase 3: Reason About Layout Completeness

Using the Layout Strategy Map from Phase 1 and your domain knowledge, check for what's missing — not just what's wrong.

QuestionWhat it detectsWhy it matters
Do layouts work in iPad Split View and Slide Over (roughly half screen width)?Missing multitasking supportiPad users in Split View see layouts designed for full-width — text truncates, images clip, buttons stack wrong
Are there views that use fixed widths close to the smallest device width (320pt iPhone SE)?Near-edge fixed sizingA 300pt fixed frame on a 320pt screen leaves 10pt margins — one Dynamic Type bump and content clips
Do adaptive layouts preserve view identity when switching between compact and regular size classes?Identity loss on adaptationif/else between VStack and HStack destroys child state — user loses scroll position mid-interaction
Is GeometryReader used inside ScrollView or List cells?GeometryReader in scrolling contextA GeometryReader in a scrolling context is proposed an unspecified height and falls back to its 10×10 ideal size, so it renders all but invisibly — constrain it with .frame(...) or use onGeometryChange
Are there layouts that assume a single window size (no Stage Manager, no free-form windows)?Missing resizable-window supportiOS 26 introduced free-form resizing; at 27 every window resizes, iPhone included — layouts that assume fixed dimensions will break
Does the app handle landscape orientation on iPhone, or only portrait?Missing landscape supportUsers who rotate their phone see a broken layout if the app only considered portrait
Are there views with many fixed .frame() calls that could use flexible alternatives?Over-constrained layoutFixed dimensions fight SwiftUI's flexible layout system — harder to maintain, more breakage

Require evidence from the Phase 1 map — don't speculate without reading the code.

Phase 4: Cross-Reference Findings

Bump severity for these combinations:

Finding A+ Finding B= CompoundSeverity
GeometryReader in stackInside ScrollView/ListCollapses to its 10×10 ideal size — near-invisibleCRITICAL
UIScreen.main.boundsUsed for layout decisionsStale values break multitaskingCRITICAL
Conditional VStack/HStackIn main content viewUser loses state on rotation/resizeCRITICAL
Large fixed frame (>300pt)No size class checkingClips on iPhone SE and iPad Split ViewHIGH
Hardcoded breakpointsDifferent values in different filesInconsistent adaptation thresholdsHIGH
Nested GeometryReadersIn frequently visited screenConfusing layout on the most-seen viewHIGH
No size class usageiPad target in deployment infoiPad users get phone-style layoutHIGH
Size class as orientation proxyiPhone Pro Max userWrong layout in landscape on large iPhoneMEDIUM

Also note overlaps with other auditors:

  • UIScreen.main / UIRequiresFullScreen → co-detected by resize-auditor (UIKit/plist/scene-manifest side; run both on mixed codebases, dedupe by file:line)
  • Non-lazy ForEach → compound with swiftui-performance-analyzer (launch lag)
  • GeometryReader in List cells → compound with swiftui-performance-analyzer (double layout pass)
  • Fixed dimensions + Dynamic Type → compound with accessibility-auditor (text clipping)
  • Missing adaptivity + iPad → compound with ux-flow-auditor (broken user journey on iPad)

Phase 5: Layout Health Score

markdown
## Layout Health Score

| Metric | Value |
|--------|-------|
| Adaptivity coverage | Size class usage: yes/no, ViewThatFits: N usages, AnyLayout: N usages |
| GeometryReader discipline | N total, M constrained with .frame() (Z%), nested: N files |
| Fixed dimension risk | N fixed frames >300pt, M hardcoded breakpoints |
| Deprecated API usage | N UIScreen/UIDevice references |
| Identity safety | N conditional stack switches, M using AnyLayout (Z% safe) |
| Device coverage | Smallest supported width: Xpt, multitasking support: yes/no |
| **Health** | **ADAPTIVE / RIGID / BROKEN** |

Scoring:

  • ADAPTIVE: No CRITICAL issues, size class or ViewThatFits used for adaptation, 0 deprecated APIs, no identity-losing conditional stacks, supports multitasking
  • RIGID: No CRITICAL issues, but missing adaptivity (no size class usage), or some fixed dimensions that risk clipping, or conditional stacks without AnyLayout
  • BROKEN: Any CRITICAL issues (GeometryReader in stacks, deprecated APIs, UIRequiresFullScreen), or layouts that clip on common device sizes

Output Format

markdown
# SwiftUI Layout Audit Results

## Layout Strategy Map
[8-10 line summary from Phase 1]

## Summary
- CRITICAL: [N] issues
- HIGH: [N] issues
- MEDIUM: [N] issues
- LOW: [N] issues
- Phase 2 (anti-pattern detection): [N] issues
- Phase 3 (completeness reasoning): [N] issues
- Phase 4 (compound findings): [N] issues

## Layout Health Score
[Phase 5 table]

## Issues by Severity

### [SEVERITY] [Category]: [Description]
**File**: path/to/file.swift:line
**Phase**: [2: Detection | 3: Completeness | 4: Compound]
**Issue**: What's wrong or missing
**Impact**: What breaks and on which devices
**Fix**: Code example showing the fix
**Cross-Auditor Notes**: [if overlapping with another auditor]

## Recommendations
1. [Immediate actions — CRITICAL fixes (GeometryReader, deprecated APIs)]
2. [Short-term — HIGH fixes (identity loss, adaptivity)]
3. [Long-term — architectural improvements from Phase 3 findings]
4. [Test on: iPhone SE (320pt), iPad Split View (~half width), iPad Stage Manager]

Output Limits

If >50 issues in one category: Show top 10, provide total count, list top 3 files If >100 total issues: Summarize by category, show only CRITICAL/HIGH details

False Positives (Not Issues)

  • GeometryReader as root view of a screen (no siblings to collapse)
  • UIScreen.main used only for one-time setup (e.g., launch screen)
  • UIRequiresFullScreen in games that want discrete snap-resizing (its one remaining role at 27)
  • Small fixed frames (<100pt) for icons/badges
  • VStack { ForEach } with <20 items (lazy overhead not worth it)
  • Size class checks that genuinely adapt layout (not inferring orientation)
  • GeometryReader with .frame() constraint (already safe)
  • Large fixed frames for full-screen backgrounds/images (intentional)
  • .ignoresSafeArea() on backgrounds, full-bleed images, maps, or video players
  • .safeAreaInset hosting content controls — media transport (play/pause/skip), a scrubber, a text composer, or a banner — rather than action buttons

Related

For SwiftUI layout patterns, reference, and containers: axiom-swiftui skill (layout)

Invocation Examples

Prompts that should launch this agent:

Explicit command: Users can also invoke this agent directly with axiom-audit-swiftui-layout

Scope

Automatically scans SwiftUI layout code for the 12 most critical violations - GeometryReader misuse, deprecated screen APIs, hardcoded breakpoints, identity loss from conditional stacks, missing lazy containers, controls under ignoresSafeArea, and hand-built toolbar rows - prevents layout breaks across device sizes, orientations, and iPhone Duo poses.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Axiom Audit Swiftui Layout AI skill do?

Use when the user mentions SwiftUI layout review, adaptive layout issues, GeometryReader problems, or multi-device layout checking.

Why use Axiom Audit Swiftui Layout on TypingMind?

Because you install it once and use it with any model. Axiom Audit Swiftui Layout is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Axiom Audit Swiftui Layout in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/CharlesWiltgen/Axiom/tree/main/axiom-codex/skills/axiom-audit-swiftui-layout. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Axiom Audit Swiftui Layout?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Axiom Audit Swiftui Layout?

As many as you like. As long as a model supports skills, you can use Axiom Audit Swiftui Layout with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Axiom Audit Swiftui Layout AI skill free?

Yes. It is published on GitHub by CharlesWiltgen under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇