Introduction

Build production-ready AI agent CLIs in minutes with the Agent Workshop.

What is Agent Workshop?

Agent Workshop is a no-code web application that transforms your ideas into fully-functional AI agent command-line interfaces. It generates complete TypeScript projects that you can customize, extend, and deploy.

The generated agents support two major AI providers:

  • Claude Agent SDK - Anthropic's official agent framework
  • OpenAI Agents SDK - OpenAI's official agent framework

Key Features

Rapid Development

Go from idea to working agent in minutes with our 7-step wizard.

Security-First

Built-in permission system with restrictive, balanced, and permissive modes.

15 Built-in Tools

File operations, web search, database queries, and more out of the box.

MCP Integration

Connect to external services via Model Context Protocol servers.

How It Works

The Agent Workshop guides you through a 7-step configuration wizard:

  1. Domain Selection - Choose your agent's area of expertise (Development, Business, Creative, Data, or Knowledge)
  2. Template Selection - Pick a pre-built template or start from scratch
  3. SDK Configuration - Select your AI provider (Claude or OpenAI) and model
  4. Tool Configuration - Enable the capabilities your agent needs
  5. MCP Configuration - Connect external servers via Model Context Protocol
  6. Project Settings - Configure metadata like name, version, and license
  7. Preview & Generate - Review and download your complete project

What You Get

After completing the wizard, you download a ZIP file containing:

1my-agent/
2├── package.json          # Dependencies and scripts
3├── tsconfig.json         # TypeScript configuration
4├── src/
5│   ├── cli.ts           # CLI entry point
6│   ├── agent.ts         # Agent logic and tool setup
7│   ├── config.ts        # Configuration management
8│   ├── permissions.ts   # Permission system
9│   └── mcp-config.ts    # MCP server setup
10├── .commands/            # Workflow commands (domain-specific)
11├── .mcp.json            # MCP server configuration
12├── README.md            # Setup and usage docs
13├── .env.example         # Environment template
14└── .gitignore           # Git ignore rules

Ready to build?

Head to the Quick Start guide to create your first agent in under 5 minutes.

Who Is This For?

  • Developers who want to quickly prototype AI-powered tools
  • Researchers who need automated literature review and data analysis
  • Business professionals automating document processing and reporting
  • Hobbyists exploring what's possible with AI agents