- Home
- Skills
- Mobile Development
- Native Unslop
More screenshots
Works with the AI Agents you already use
Native Unslop
Eliminate generic AI UI slop in native iOS (SwiftUI/UIKit) and Android (Jetpack Compose/M3) apps with token locking, a Design Intent Brief gate, and platform-native checks.
$5
Native Unslop
Example session with this skill installed
Build a native SwiftUI transfer confirmation and portfolio summary screen for our existing iOS finance app.
- Read your context and instructions
- Compiled the native unslop
Runs scripts/scan_ui_tokens.py to lock onto repo Color/Font tokens and detect the High-Trust Finance archetype, outputs the mandatory 4-point Design Intent Brief before code, loads only references/ios-swiftui-uikit.md, and delivers SwiftUI views using .glassEffect(_:in:), .insetGrouped List, .monospacedDigit(), .sheet(presentationDetents:), .sensoryFeedback, ViewThatFits XXL reflow, and Loading/Empty/Error states.
Connects securely to your tools. The creator never sees your data.
About this skill
Native Unslop eliminates generic AI UI slop in native iOS (SwiftUI, UIKit) and Android (Jetpack Compose, Material 3) applications—treating handheld software as a quiet, precise instrument rather than an advertising billboard.
4-Phase Execution Protocol
- Phase 0 — Strict Negative Boundary: Immediately halts on React Native, Flutter (
pubspec.yaml), or Web/Tailwind repos so cross-platform abstractions never contaminate native code. - Phase 1 — Context & Archetype Discovery: Runs the zero-dependency
scripts/scan_ui_tokens.pyAST/regex scanner to extract repoColor,Font, spacing,ButtonStyle,ViewModifier, and architecture tokens (@Observable, TCA,ViewModel/StateFlow) and classify one of 5 native archetypes (Clean Utility, High-Trust Finance, Editorial & Warm, Punchy & Energetic, Media & Immersive). Locks onto discovered repo tokens (locked_to_repo) or triggers a 2-option Alignment Fork on greenfield/ambiguous apps. - Phase 2 — Mandatory Design Intent Brief (Hard Gate): Forbids emitting UI code until stating the 4-point brief (Archetype & Tone, 60/30/10 Color/Surface/WCAG AA Contrast, Layout Metaphor & Hero Anchor ≤35%, Tactile & State Honesty).
- Phase 3 — Single-Platform Reference Load: Loads only
references/ios-swiftui-uikit.mdorreferences/android-compose.md(never both) to preserve context budget and keep platform idioms pure. - Phase 4 — Anti-Slop Finish Gate: Enforces
references/slop-taxonomy-native.mdandreferences/native-ux-psychology.md:- Resolves the Card Fallacy (allows singular aggregate entity cards; bans card-in-card row wrapping)
- Enforces native SwiftUI Liquid Glass (
.glassEffect(_:in:)) and.ultraThinMaterial—no fake gradient/specular overlays - Enforces WCAG AA contrast (≥4.5:1 text, ≥3:1 large/icons), dark-mode OLED surface borders, and strictly bans "ghost text" (
.tertiary/.quaternaryon copy) - Enforces 30/60/10 viewport budgeting, ≤35% hero height ceiling, peeking fold cutoffs, and physical aspect ratios (
1.586:1cards,16:9media) - Enforces platform spring physics, semantic
.sensoryFeedback/LocalHapticFeedback, ≥44pt / ≥48dp hit targets, and all 4 native states (Loading, Empty, Error, XXL Font Reflow).
Includes 5 deep reference guides, the Python token scanner, and 6 architectural case-study audits (Onyx Vault, Kuro Atelier, Pulse Live, Kyoto Expedition, Apex Portfolio, and Daily Habit Tracker), paired with Before vs. After visual proofs at https://jozol.github.io/true-native/.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Mobile Development.
Does not support React Native, Flutter, Ionic, or web/Tailwind projects (Phase 0 stops execution if cross-platform markers are detected). Token scanning uses regex heuristics rather than full compiler AST parsing. Full Swift/Kotlin example source files are omitted to keep the skill package lean; architectural audits are bundled under examples/*/README.md and Before/After visual comparisons are hosted at https://jozol.github.io/true-native/#case-studies.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed today
- Software
Python 3.8+
iOSorAndroid codebase- Optional
- Xcode (for iOS builds)Android Studio (for Compose)
Zero third-party dependencies. The bundled token scanner uses Python 3.8+ standard library only.
Permissions
File Scopes
Runs locally with zero network requests and zero third-party Python dependencies.