Meet-Miyani/compose-skill

Description: AI agent skill for Jetpack Compose & Compose Multiplatform (KMP/CMP). MVI architecture, Navigation 3, Koin/Hilt, Ktor, Room, DataStore, Paging 3, Coil, coroutines/Flow, animations, performance, accessibility, testing, and cross-platform patterns. Works with Codex, Cursor, Claude Code.

<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> This project provides an AI agent skill for Jetpack Compose and Compose Multiplatform (KMP/CMP), enabling coding assistants like Codex, Cursor, and Claude Code to understand production-grade development practices. It addresses AI agents' limitations in grasping complex UI frameworks, offering expertise in architecture, state management, UI, networking, persistence, performance, and accessibility to accelerate development and improve code quality.

How It Works

<2-4 sentences on core approach / design (key algorithms, models, data flow, or architectural choices) and why this approach is advantageous or novel.> The skill functions as a knowledge base for AI agents, consisting of a SKILL.md definition and 37 detailed Markdown reference files. It guides agents to apply MVI architecture, unidirectional data flow, and effective state modeling (StateFlow, Channel). This empowers AI agents to generate, refactor, and review Compose code with architectural understanding, moving beyond mere syntax.

Quick Start & Requirements (only include this section if it contains useful information)

• Primary install / run command (pip, Docker, binary, etc.).
• Non-default prerequisites and dependencies (GPU, CUDA >= 12, Python 3.12, large dataset, API keys, OS, hardware, etc.).
• Estimated setup time or resource footprint.
• If they are present, include links to official quick-start, docs, demo, or other relevant pages. Installation involves cloning the repository into the AI agent's skill directory (e.g., ~/.cursor/skills/compose-skill). The agent requires SKILL.md, agents/, and references/. No runtime dependencies are installed. Refer to your AI agent's documentation for specific installation and verification steps.

Highlighted Details

• Bullet 1 (benchmarks, performance claims, novel integration, etc.)
• Bullet 2
• Bullet 3
• Bullet 4 (optional)

• Enhances AI agents (Codex, Cursor, Claude Code) with Compose and KMP/CMP expertise.
• Covers MVI architecture, state management (StateFlow, Channel), and unidirectional data flow.
• Integrates modern libraries: Navigation 3, Ktor, Room, DataStore, Paging 3, Coil 3.
• Focuses on performance optimization, accessibility (WCAG), and cross-platform patterns (commonMain, expect/actual).
• Features 37+ deep-dive reference files on diverse topics, from coroutines to CI/CD.

Maintenance & Community

• Notable contributors, sponsorships, partnerships, deprecations, migrations, or other health signals if notable.
• Links to Discord/Slack, social handles, roadmap, etc. Contributions are welcomed via standard Git workflows. The repository includes validation scripts. However, the README does not list specific contributors, sponsorships, community channels (Discord/Slack), or a public roadmap.

Licensing & Compatibility

• License type and notable restrictions (GPL -> copyleft, SSPL, etc.).
• Compatibility notes for commercial use or closed-source linking. Licensed under the permissive MIT License, allowing for commercial use and integration into closed-source projects. Compatible with AI Skills-standard tools, including Codex, Cursor, and Claude Code.

Limitations & Caveats

<1-3 sentences on caveats: unsupported platforms, missing features, alpha status, known bugs, breaking changes, bus factor, deprecation, etc. Avoid vague non-statements and judgments.> Effectiveness relies on the AI agent's interpretation of the knowledge base. The README does not detail specific bugs or unsupported platforms beyond KMP scope. Users should note that AI agent skill paths may change.