Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Thalarch Compose Ui

ASecurity

Product-quality Jetpack Compose UI workflow for redesigning or extending an existing Android app without losing its design language, accessibility, adaptive behavior, localization, state correctness, or runtime performance. Use for Compose screens, settings, media/player UI, visual redesigns, or interaction-heavy Android surfaces that require rendered device evidence.

2 stars
0 votes
0 copies
1 views
Added 9/19/2026
ai-agentsexpressrailsperformance

Security Analysis

A100/100

Scanned 9/19/2026

$npx -y skills add LUC4N3X/antigravity-thalarch --skill thalarch-compose-ui --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Thalarch Compose Ui?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Thalarch Compose Ui
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/luc4n3x-thalarch-compose-ui/badge)](https://www.skillsdirectory.com/skills/luc4n3x-thalarch-compose-ui)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: thalarch-compose-ui
description: >
  Product-quality Jetpack Compose UI workflow for redesigning or extending an existing Android app
  without losing its design language, accessibility, adaptive behavior, localization, state
  correctness, or runtime performance. Use for Compose screens, settings, media/player UI, visual
  redesigns, or interaction-heavy Android surfaces that require rendered device evidence.
---

# Thalarch Compose UI

A Compose screen is both a rendered product surface and a stateful runtime system. Treat both as
acceptance criteria.

## 1. Extract before inventing

Before redesigning an existing app, inspect:

- theme/color/typography tokens;
- reusable components;
- spacing/radius/divider patterns;
- navigation and top-level screen structure;
- dark/light/dynamic-color behavior;
- screenshots or supplied references;
- wording/localization conventions.

Use `thalarch-design-system` when the visual language is unclear or the redesign is substantial.
Preserve stable product identity unless the user explicitly asks to change it.

## 2. Write a compact visual contract

For a meaningful redesign, establish:

- hierarchy;
- density;
- grouping;
- primary interaction;
- one distinctive product-native visual idea;
- responsive rules;
- accessibility constraints;
- explicit anti-patterns.

Do not implement a pile of independently styled cards and call it a design system.

## 3. State and recomposition discipline

Inspect:

- state ownership and hoisting;
- stable keys in lazy content;
- `remember` / `rememberSaveable` lifetime;
- `derivedStateOf` only when derivation cost/recomposition semantics justify it;
- effects and lifecycle ownership;
- expensive mapping/sorting/parsing inside composables;
- allocation-heavy objects created during frequent recomposition.

A visually correct screenshot does not prove state correctness or performance.

## 4. Interaction quality

Controls should expose their actual state clearly and behave consistently.

Check:

- enabled/disabled/read-only states;
- pressed/focused/selected feedback;
- slider/drag gesture precision;
- reset/default actions;
- destructive actions and confirmation where appropriate;
- scroll behavior and nested gestures;
- keyboard/input behavior when applicable.

Prefer platform/material primitives when they already express the interaction correctly. Custom
components need a product or interaction reason, not novelty.

## 5. Accessibility

Verify, where relevant:

- touch targets of at least the platform-appropriate minimum (normally 48dp on Android);
- meaningful semantics/content descriptions;
- TalkBack order and labels;
- role/state/value semantics for sliders, toggles, tabs, and custom controls;
- contrast;
- font scaling;
- focusability;
- RTL mirroring and AutoMirrored icons.

Decorative assets should not create noisy accessibility output.

## 6. Adaptive layout

Test the actual surfaces the product supports:

- small phone;
- typical phone;
- large phone/tablet when relevant;
- portrait/landscape;
- edge-to-edge/insets;
- font scaling;
- navigation bars/cutouts.

Do not treat landscape as a wider screenshot of the portrait composition when the layout contract
requires reflow.

## 7. Localization is part of layout

Combine with `thalarch-localization` when new or changed user-facing strings exist.

Design for:

- longer translated labels;
- plural/placeholder expansion;
- RTL;
- line wrapping/truncation;
- technical terms that should remain standard rather than be awkwardly literalized.

A default-English render is not proof that the screen works across supported locales.

## 8. Professional settings/product language

Settings should describe capability, state, and consequence—not advertise themselves unless the
surface is genuinely a marketing/paywall surface.

Avoid invented prestige labels such as `Premium Engine`, `Ultimate`, `Pro Processing`, or similar
marketing copy when they do not correspond to a real product tier or user-visible distinction.
Prefer concise domain language such as `Audio`, `Equalizer`, `Playback`, `Processing`, or the
project's established terminology.

## 9. Anti-template / anti-AI guardrails

Do not add by default:

- random glassmorphism;
- neon/purple glow;
- gradients without hierarchy purpose;
- cards inside cards inside cards;
- oversized rounded containers everywhere;
- decorative pills for ordinary labels;
- animation on every interaction;
- generic dashboard visual language inside a media/product screen.

Distinctive does not mean busy.

## 10. Motion

Use motion for continuity, state transition, or feedback. Keep it restrained and respect reduced
motion where the platform/project supports it.

Never animate a frequently changing value in a way that creates unnecessary recomposition or input
lag.

## 11. Rendered proof loop

For visually consequential changes:

1. inspect existing rendered UI when available;
2. implement the smallest coherent slice;
3. build;
4. launch on emulator/device;
5. capture/inspect the actual rendered screen;
6. test interaction and state changes;
7. inspect at least the required theme/viewport variants;
8. adjust from rendered evidence, not source optimism.

Source code is not visual proof. Preview-only output is weaker evidence than the runtime surface when
runtime behavior is part of acceptance.

## 12. Completion

Report separately:

- source/build correctness;
- interaction/state evidence;
- accessibility checks;
- localization coverage;
- rendered/device evidence;
- any variants that remain `UNVERIFIED`.

Attribution

LUC4N3XLUC4N3X
View sourceSee grades on GitHubMore from LUC4N3X →
SSkills Directory ProSkills Directory

Get any skill into Claude in one click.

Download any skill as a ZIP for Claude.ai, Claude Desktop, or .claude/skills. $9/mo.

See Pro

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills Directory ProSkills Directory

Get any skill into Claude in one click.

Download any skill as a ZIP for Claude.ai, Claude Desktop, or .claude/skills. $9/mo.

See Pro

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1087401 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

697551 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →