Beginner · ~3 minutes

Quick Start

Three steps. That's it. Initialize a project, open your IDE, and tell your AI agent what to build.

1 Install & Initialize

Install the CLI and create a project:

npm install -g @prodara/cli
prodara init my-app
cd my-app

That's all the CLI does — it scaffolds a .prd spec file and configures your AI agent automatically. The Prodara VS Code extension (optional) gives you syntax highlighting and real-time diagnostics.

💡
5 templates are available: minimal, saas, marketplace, internal-tool, and api. Use --template saas to start with a richer spec.

2 Open Your IDE & Build

Open the project and tell your agent what to build:

code my-app

In Copilot Chat (or any AI agent), just say:

Copilot Chat
/Prodara Build me a ToDo application with tasks,
due dates, and priority levels.

Your agent handles everything — it writes the .prd spec, runs the compiler, and implements the code. No need to learn any commands or syntax. The agent already knows how Prodara works.

ℹ️
prodara init generates an @prodara prompt and copilot-instructions automatically. Your agent is ready to build from the moment you open the project.

3 Iterate

Want to change something? Just ask:

Copilot Chat
/Prodara Add a "completed" filter and a
workflow to archive old tasks.

The agent updates the spec, rebuilds, and implements the changes. The VS Code extension shows real-time validation so you and your agent catch errors instantly.

? Optional: CLI Reference

Most users never need these — your agent runs them for you. But if you want to run commands manually: 

prodara build          # Full pipeline
prodara validate       # Type-check .prd files
prodara test .         # Run spec tests
prodara diff           # Show what changed

What's Next?

You've built your first Prodara project! Here's where to go next: