Claude Code Development Kit

A lightweight starter kit for Claude Code subscribers. Gives your project a solid foundation — documentation structure, code review automation, image tools, and sensible defaults — that you extend as you go.

This isn't a heavy framework. It's the setup you'd build yourself after a few weeks of using Claude Code, packaged so you start with it on day one.

Who This Is For

You have a Claude Code subscription and want to hit the ground running. Maybe you're building an app, a side project, or starting something new. You want Claude to understand your project structure, review its own work, and not accidentally git push --force your main branch.

This kit gives you that. Install what you need, skip what you don't, extend it however you want.

Quick Start

One command:

curl -fsSL https://raw.githubusercontent.com/peterkrueck/Claude-Code-Development-Kit/main/install.sh | bash

Or clone and run:

git clone https://github.com/peterkrueck/Claude-Code-Development-Kit.git
cd Claude-Code-Development-Kit
./setup.sh

The installer walks you through what to include. Everything is optional except the core.

What You Get

Always installed (core)

• CLAUDE.md — A template for your project's AI instruction set. This is how you tell Claude your project's rules, architecture decisions, and constraints.
• /prime command — Loads your documentation into context. Run it at the start of a session.
• /update-docs skill — Keeps your documentation in sync after code changes.
• Documentation scaffolding — Four structured files (spec.md, project-structure.md, progress.md, deployment-infrastructure.md) plus directories for legal, business, design, and open issues.
• Security scanner — Blocks API keys and secrets from leaking through MCP plugins.
• Deny list — Prevents git push --force, rm -rf, git reset --hard, and other destructive commands.
• Asset directories — assets/ for your app icons, logos, and graphics.

Optional: Review skills

Independent code review using Google's Gemini CLI — a completely different AI architecture that catches things Claude might miss. When Gemini is unavailable, falls back to a Claude sub-agent.

• /review-work — Sends your uncommitted diff to Gemini with a review checklist
• /second-opinion — Auto-triggers when Claude faces tricky architecture decisions

Requires: Gemini CLI

Optional: Visual skills

Generate app icons, character art, social media graphics, and marketing assets. Edit images with precision. Remove backgrounds locally.

• /image-gen — AI image generation via Nanobanana 2 (Gemini). Optional reference photos for style consistency.
• /image-edit — Crop, resize, rotate, mirror. Analyzes content bounds before cutting.
• /bg-remove — Background removal via rembg. Runs locally, nothing sent externally.

Requires: Python 3, Pillow, numpy. /image-gen also needs a GEMINI_API_KEY. /bg-remove needs rembg.

Optional: Deploy skill template

A customizable deployment pipeline you fill in with your own commands. Follows: detect changes → run tests → deploy → verify → report.

Optional: Gemini integration

GEMINI.md — an instruction file that Gemini CLI reads automatically. Gives Gemini full context about your project when invoked as a reviewer or consultant.

Optional: Review-on-stop hook

Nudges you to review before finishing. When you have 10+ lines of new code (net changes this session, not pre-existing dirty state), the first stop shows an advisory with changed files and suggests review/test/docs. Stop again for a reminder, a third time to skip entirely. Never traps you — three stops always gets you out.

Optional: Audio notifications

Plays a sound when Claude finishes a task or needs your input. Supports macOS and Linux.

What Gets Installed

your-project/
├── .claude/
│   ├── commands/
│   │   └── prime.md                    # /prime — load project context
│   ├── hooks/
│   │   ├── security-scan.sh            # Blocks secrets from leaking to plugins
│   │   ├── review-on-stop.sh           # Advisory review nudge (if selected)
│   │   ├── snapshot-baseline.sh        # Session baseline capture (if selected)
│   │   ├── notify.sh                   # Audio notifications (if selected)
│   │   ├── config/
│   │   │   ├── pipeline.json           # Review-on-stop configuration
│   │   │   └── sensitive-patterns.json # Security scan patterns
│   │   └── sounds/
│   │       ├── complete.wav
│   │       └── input-needed.wav
│   ├── skills/                         # Selected skills (review, visual, deploy)
│   └── settings.local.json             # Permissions, hooks, deny list
│
├── assets/                             # Your visual assets
│
├── docs/
│   ├── ai-context/
│   │   ├── spec.md                     # What the product does
│   │   ├── project-structure.md        # File tree and tech stack
│   │   ├── progress.md                 # Roadmap and task tracking
│   │   └── deployment-infrastructure.md # Hosting, secrets, CI/CD
│   ├── legal/
│   ├── business/
│   ├── design-brand/
│   └── open-issues/
│
├── CLAUDE.md                           # Your project's AI rules
└── GEMINI.md                           # Gemini instructions (if selected)

Upgrading from v2

v3 is a major rewrite. Commands became skills, the 3-tier doc system became 4 focused files, Gemini moved from MCP to native CLI, Context7 is now a plugin.

v2 is still available:

curl -fsSL https://raw.githubusercontent.com/peterkrueck/Claude-Code-Development-Kit/v2/install.sh | bash

See CHANGELOG.md for the full migration guide.

FAQ

Do I need Gemini CLI? Only for the review skills. Everything else works without it.

Can I use this with Cursor/Windsurf/Codex? The skills and hooks are Claude Code-specific. The documentation templates work with any AI tool. SKILLS should work with most AI coding tools by now.

Is this for large teams? It works for teams, but it's designed for individual developers and small teams. If you're running complex multi-agent workflows at scale via the API, this probably isn't what you need.

License

MIT — see LICENSE.