Kilo Code Kit

Kilo Code Kit provides a modular AI development system built around Agents, Skills, and Workflows.

It installs a ready-to-use AI architecture that automatically selects specialist agents and loads the correct skills for each task.

Features

• Automatic AI agent routing
• Modular skills system
• Reusable workflows
• Security and coding standards
• Designed for Kilo Code environments

Quick Install

Run:

npx kilo-code-kit init

This installs the following into your project:

• .kilo/
• .kilocode/
• scripts/
• AGENTS.md
• GEMINI.md
• kilo.json

Project Structure

After installation your project will contain:

.kilo/
  agents/

.kilocode/
  skills/
  workflows/
  rules/

scripts/

AGENTS.md
GEMINI.md
kilo.json

Agents

Agents are specialist AI personas.

Each agent focuses on a specific domain.

Examples include:

• Frontend Specialist
• Backend Specialist
• Security Auditor
• Debugger
• Project Planner

Agents automatically activate when the system detects a relevant request.

Example:

User: Add JWT authentication

AI: 🤖 Applying @security-auditor + @backend-specialist

Agents are stored in:

.kilo/agents/

Each agent defines:

• role
• expertise
• behavior rules

Skills

Skills are knowledge modules that agents load when needed.

Skills provide domain expertise such as:

• clean code practices
• API design patterns
• database architecture
• testing strategies
• performance optimization

Skills are stored in:

.kilocode/skills/

Each skill contains a SKILL.md file that describes:

• best practices
• guidelines
• implementation patterns

Agents read skill descriptions and load them automatically when tasks match their domain.

Workflows

Workflows automate common development tasks.

They define step-by-step procedures for solving problems.

Examples include:

• brainstorming ideas
• creating new features
• debugging issues
• running tests
• deploying applications

Workflows can be triggered using slash commands.

Example:

/create authentication system
/debug login error
/test API endpoints

Workflows are stored in:

.kilocode/workflows/

How Automatic Agent Routing Works

The system automatically detects the best agent for a request.

Steps:

1. Analyze the user request
2. Detect relevant domains (frontend, backend, security, etc.)
3. Select specialist agents
4. Load relevant skills
5. Generate expert-level responses

This allows developers to simply describe their task without manually selecting agents.

Example:

User: Fix the login button

AI: 🤖 Applying @frontend-specialist

CLI Tool

The toolkit includes a CLI installer.

Command:

npx kilo-code-kit init

This copies the AI configuration files into your project.

Option: --force

Example:

npx kilo-code-kit init --force

This overwrites existing files.

Safe Installation

The installer automatically protects existing configurations. If .kilo or .kilocode folders already exist in your project, the installer will warn you and skip the installation.

Example output:

Installing Kilo Code Kit...

⚠ .kilo already exists in the target project
⚠ .kilocode already exists in the target project

⚠ Kilo Code configuration already exists
Use --force to overwrite existing files

License

MIT License

Contributing

Contributions are welcome.

You can add:

• new agents
• new skills
• new workflows

to expand the toolkit.