aldefy/compose-skill

Jetpack Compose Agent Skill — AI-powered coding guidance with actual androidx/androidx source code receipts. Works with Claude Code, Codex CLI, Gemini CLI, Cursor, Copilot, Windsurf, and more.

What it solves

AI coding assistants often generate Jetpack Compose code that compiles but contains subtle bugs, such as incorrect state management (remember vs derivedStateOf), unstable recompositions, broken modifier ordering, and hallucinated APIs. This project provides a structured knowledge base that prevents these "vibes-based" guesses by giving AI agents access to verified reference guides and actual source code.

How it works

The project is distributed as a set of Markdown files that can be integrated into various AI coding tools (like Claude Code, Cursor, Copilot, and Windsurf) as a "skill" or rule. It operates in two layers:

  1. Guidance Docs: 24 reference guides covering specific Compose topics (e.g., state management, performance, navigation, and multiplatform) that map task signals to specific patterns and pitfalls.
  2. Source Code Receipts: 6 files containing actual Kotlin source code from androidx/androidx and compose-multiplatform-core, allowing the agent to verify implementation details instead of guessing.

Who it’s for

Android and Multiplatform developers using AI coding assistants who want to ensure the generated UI code follows production-ready patterns and avoids common Compose performance and stability pitfalls.

Highlights

  • Comprehensive Coverage: Includes specialized guides for Android TV, Material 3 motion, Paging 3, and Nav 2 to Nav 3 migration.
  • Source-Backed: Uses real source code from official repositories to eliminate API hallucinations.
  • Multi-Tool Support: Ready-to-use installation instructions for Claude Code, Cursor, Copilot, Gemini CLI, Windsurf, and Amazon Q.
  • Production Focused: Includes a "production crash playbook" with root causes and fixes for common real-world crash patterns.

Related

  • Project
    codeaholicguy/ai-devkit

    AI DevKit is a local npm‑based control plane that unifies configuration, monitoring, cross‑agent messaging, searchable SQLite memory, and composable engineering “skills” (requirements, design, testing, review, etc.) for many AI coding assistants such as Claude Code, Gemini CLI, Codex CLI, Cursor, and GitHub Copilot. It lets teams manage multiple agents from a single console, keep project conventions in a searchable store, and enforce a senior‑engineer workflow without modifying the underlying LLMs.

  • Project
    NeoLabHQ/context-engineering-kit

    A plug‑in marketplace for LLM‑based coding assistants (Claude Code, Gemini, Antigravity, etc.) that supplies token‑efficient, context‑engineering patterns—reflection loops, sub‑agent judges, specification‑first development, code‑review agents, and Git helpers—to make AI‑generated code more reliable and easier to integrate into CI pipelines.

  • Project
    firebase/agent-skills

    A collection of packaged instructions and scripts that extend AI coding agents' ability to understand and work with Firebase.

  • Project
    BuilderIO/skills

    A collection of composable skills for coding agents that improve their autonomy, visual planning, and efficiency in software development tasks.

  • Project
    caliber-ai-org/ai-setup

    A tool that automatically generates and maintains AI agent context files to prevent hallucinations and keep AI coding assistants in sync with codebase changes.