Getting Started — Camel-Kit

Overview

Camel-Kit is an AI-guided integration development tool that transforms how you build Apache Camel integrations. Instead of writing code directly, you describe your integration requirements in natural language, and the AI handles design, planning, implementation, and verification.

Installation Walkthrough

Follow these steps to install Camel-Kit and create your first integration project.

Prerequisites

Before installing Camel-Kit, ensure you have:

Java 17 or higher - Required for running Apache Camel and JBang
JBang 0.120.0 or higher - The scripting engine that powers Camel-Kit

Verify Java:

java -version

Install JBang:

Visit jbang.dev or use your package manager:

# macOS
brew install jbangdev/tap/jbang

# Linux
curl -Ls https://sh.jbang.dev | bash -s - app setup

# Windows
choco install jbang

Install Camel-Kit

Method 1: JBang App Install (Recommended)

The easiest way to install Camel-Kit:

jbang app install camel-kit@luigidemasi/camel-kit

This adds the camel-kit command to your PATH.

Verify installation:

camel-kit --version

Alternative Methods:

Run without installing: jbang camel-kit@luigidemasi/camel-kit init ...
Camel JBang plugin: camel plugin add camel-kit

Initialize Project

Initialize Your First Project

Create a new integration project:

camel-kit init my-project --ai claude
cd my-project

What gets created:

AGENTS.md - Routing table for AI agents
.claude/commands/ - Pipeline phase slash commands
.camel-kit/ - Governance documents (constitution, iron-laws)
.mcp.json - Catalog integration config

Choose your AI agent:

Agent	Flag
Claude Code	`--ai claude` (recommended)
IBM Bob	`--ai bob`
Gemini CLI	`--ai gemini`
Qwen	`--ai qwen`
OpenCode	`--ai opencode`

First Command

Run Your First Command

For greenfield development:

Open your AI coding agent in the project directory and say:

I want to build an integration that reads from Kafka and writes to a database

The AI reads AGENTS.md, invokes /camel-brainstorm, and guides you through design.

For migration:

camel-kit init --here --ai claude --source-platform mulesoft

Then tell your AI agent:

Migrate my MuleSoft flows to Camel

The AI invokes /camel-migrate and handles the conversion.

What's Next

What’s Next?

Now that you have Camel-Kit installed and a project initialized:

Learn the pipeline:

Read the Pipeline Overview to understand the three phases

Choose your workflow:

Greenfield Workflow - Build integrations from scratch
Migration Workflow - Convert existing integrations

Explore features:

Graph CLI - Migration analysis tools
Knowledge - Q&A about Camel components

Get help:

Documentation at this site
GitHub: github.com/luigidemasi/camel-kit
Community discussions in the repository

Two Workflows

Camel-Kit supports two integration workflows. Toggle to see each approach:

Greenfield Development

Build new integrations from scratch using the Design → Plan → Execute pipeline:

Run /camel-brainstorm — AI interviews you about requirements
Review and approve the design specification
AI automatically decomposes into tasks and generates code
AI verifies the integration with runtime tests

camel-kit init my-project --ai claude
# Then in your AI agent:
# "I want to build a Kafka-to-PostgreSQL integration"

→ Greenfield Workflow guide

Platform Migration

Migrate existing integrations from other platforms:

MuleSoft Mule 3.x/4.x — Full XML and DataWeave support
Apache Camel 2.x/3.x — Modernize to Camel 4.x YAML DSL

The AI auto-detects the source platform, uses graph-based flow analysis, and converts flow-by-flow.

camel-kit init --here --ai claude --source-platform mulesoft
# Then in your AI agent:
# "Migrate my MuleSoft flows to Camel"

→ Migration Workflow guide