Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
thebushidocollective avatar

Android Jetpack Compose

  • 1.8k installs
  • 186 repo stars
  • Updated July 19, 2026
  • thebushidocollective/han

Process for creating Han discipline plugins that bundle specialized agents with defined expertise areas, responsibilities, and invocation triggers

About

This skill guides developers through creating discipline plugins within the Han marketplace that encapsulate specialized agents for Android Jetpack Compose expertise. It provides structured plugin architecture including agent definitions, configuration files, and registration processes. Developers learn to define agent roles, responsibilities, and invocation triggers using markdown templates. The discipline plugin pattern enables teams to package domain-specific knowledge as reusable agents with clear expertise boundaries. Includes color-coded agent guidelines and marketplace registration steps for seamless integration.

  • Discipline plugin architecture with agent definitions and configuration templates
  • Color-coded agent role guidelines for visual identification and team coordination
  • Step-by-step plugin creation from structure to marketplace registration
  • Agent markdown templates with responsibility areas and invocation scenarios
  • Han marketplace integration with plugin.json and marketplace.json registration

Android Jetpack Compose by the numbers

  • 1,806 all-time installs (skills.sh)
  • +51 installs in the week ending Aug 4, 2026 (Skillselion tracking)
  • Ranked #706 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
  • Security screen: MEDIUM risk (skills.sh audit)
  • Data as of Aug 4, 2026 (Skillselion catalog sync)
npx skills add https://github.com/thebushidocollective/han --skill android-jetpack-compose

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs1.8k
repo stars186
Security audit3 / 3 scanners passed
Last updatedJuly 19, 2026
Repositorythebushidocollective/han

What it does

Create specialized discipline plugins with agent definitions for Android Jetpack Compose development workflows

Who is it for?

Team leads building internal agent libraries, framework contributors extending Han ecosystem, developers standardizing expertise across projects

Skip if: Individual contributor automation, generic skill development, non-agent-based extensions

When should I use this skill?

Creating new specialized agents, packaging team expertise as plugins, extending Han marketplace with discipline-specific tools

What you get

Developers successfully create marketplace-ready discipline plugins with clear agent definitions, proper configuration, and integration with Han infrastructure

  • Compose UI components
  • State management implementations

By the numbers

  • Documents 4 Jetpack Compose state APIs: remember, rememberSaveable, mutableStateOf, and derivedStateOf

Files

SKILL.mdMarkdownGitHub ↗

Create a Discipline (道 - Do) Plugin

Create a new discipline plugin for: $ARGUMENTS

What is a Discipline Plugin?

Discipline plugins represent "the way" or path of mastery in the Han marketplace. They provide specialized agents that embody expertise in specific development disciplines. Each discipline contains one or more agents with deep domain knowledge.

Plugin Structure

Create the following directory structure:

plugins/disciplines/{discipline-name}/
├── .claude-plugin/
│   └── plugin.json          # Plugin metadata (ONLY plugin.json goes here)
├── agents/
│   └── {agent-name}.md     # Agent definition
├── han-plugin.yml           # Han configurations (at plugin root)
├── hooks/
│   └── hooks.json           # Claude Code hooks (optional)
├── skills/
│   └── {skill-name}/
│       └── SKILL.md        # Skill documentation (optional)
└── README.md               # Plugin documentation

IMPORTANT:

  • Only plugin.json goes inside .claude-plugin/
  • hooks.json goes in the hooks/ directory
  • han-plugin.yml stays at the plugin root (NOT in hooks/)
  • Discipline names should be descriptive (e.g., frontend-development, api-engineering, security-engineering)

Step 1: Create plugin.json

Create .claude-plugin/plugin.json:

{
  "name": "{discipline-name}",
  "version": "1.0.0",
  "description": "Specialized agents for {discipline area} including {key responsibilities}.",
  "author": {
    "name": "The Bushido Collective",
    "url": "https://thebushido.co"
  },
  "homepage": "https://github.com/thebushidocollective/han",
  "repository": "https://github.com/thebushidocollective/han",
  "license": "MIT",
  "keywords": [
    "{discipline}",
    "{specialty}",
    "agent",
    "{related-area}"
  ]
}

Step 2: Create han-plugin.yml

Create han-plugin.yml at the plugin root:

# {discipline-name} plugin configuration
# This plugin provides specialized agents for {discipline area}

# No hooks for agent-focused plugins (unless needed)
hooks: {}

# No MCP server (unless this discipline provides one)
mcp: null

# Memory provider (optional)
memory: null

Step 3: Create Agent Definitions

For each specialized role, create an agent markdown file.

Agent File: agents/{agent-name}.md

---
name: {agent-name}
description: |
  {One-paragraph description of the agent's expertise, focus areas, and when to invoke them.}
model: inherit
color: {purple|blue|green|red|yellow|cyan|magenta}
---

# {Agent Title}

## Role

{Detailed description of the agent's role and expertise}

## Core Responsibilities

### {Responsibility Area 1}
- {Specific capability}
- {Specific capability}

## Approach

1. **{Step 1}**: {What the agent does}
2. **{Step 2}**: {What the agent does}

## When to Invoke

- {Specific scenario}
- {Specific scenario}

Agent Color Guidelines

  • purple: Design, UI/UX, presentation
  • blue: Architecture, planning, infrastructure
  • green: Testing, quality, reliability
  • red: Security, performance, critical systems
  • yellow: Documentation, communication
  • cyan: Data, analytics, integration
  • magenta: Creative, experimental, research

Step 4: Write README.md

# {Discipline Name}

{Description of the discipline and what these agents provide}

## Specialized Agents

### {Agent 1 Name}
{Description} | **Invoke when**: {Specific scenarios}

### {Agent 2 Name}
{Description} | **Invoke when**: {Specific scenarios}

## Installation

\`\`\`bash
han plugin install {discipline-name}
\`\`\`

Step 5: Register in Marketplace

Add your plugin to .claude-plugin/marketplace.json:

{
  "name": "{discipline-name}",
  "description": "Specialized agents for {discipline area} including {key responsibilities}.",
  "source": "./plugins/disciplines/{discipline-name}",
  "category": "Discipline",
  "keywords": [
    "{discipline}",
    "{specialty}",
    "agent"
  ]
}

Naming Guidelines

Discipline names should be descriptive and specific:

GoodToo Generic
frontend-developmentfrontend
api-engineeringapi
security-engineeringsecurity
database-engineeringdatabases
site-reliability-engineeringsre

Examples of Well-Structured Discipline Plugins

  • frontend-development: Excellent agent definitions with clear roles
  • security-engineering: Strong technical depth
  • system-architecture: Good balance of strategy and implementation

Questions?

See the Han documentation or ask in GitHub Discussions.

Related skills

How it compares

Pick android-jetpack-compose over generic Kotlin skills when agents must emit recomposition-safe Jetpack Compose with correct state API selection.

FAQ

Which Compose state APIs does android-jetpack-compose cover?

android-jetpack-compose documents remember, rememberSaveable, mutableStateOf, and derivedStateOf. remember survives recomposition, rememberSaveable survives configuration changes, and derivedStateOf handles computed dependent state.

When does android-jetpack-compose activate?

android-jetpack-compose activates when building Android UIs with Jetpack Compose, managing state with remember or mutableStateOf, or implementing declarative UI patterns in Kotlin Android projects.

Is Android Jetpack Compose safe to install?

skills.sh reports 3 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.

AI & Agent Buildingfrontenddevops

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.