Getting Started with Mastra

Mastra is a framework for building applications that can integrate with Large Language Models (LLMs). You can use it to create applications that leverage agents, tools, and workflows with various LLM providers like OpenAI, Anthropic, or Groq.

Installation

There are two ways to install Mastra: automatic and manual installation.

Automatic Installation (Recommended)

You can create a new Mastra project using the create-mastra tool, which will scaffold your project with all the necessary configurations:

BASHCopied
# Using npx
npx create-mastra@latest

# Using npm
npm create mastra

# Using yarn
yarn create mastra

# Using pnpm
pnpm create mastra

During the installation, you'll be guided through several prompts:

What do you want to name your project? my-mastra-app
Choose components to install:
◯ Agents (recommended)
◯ Tools
◯ Workflows
Select default provider:
◯ OpenAI (recommended)
◯ Anthropic
◯ Groq
Would you like to include example code? No / Yes

After completing these prompts, create-mastra will set up your project directory with TypeScript, install dependencies, and configure your selected components and LLM provider.

Manual Installation

If you prefer to set up your project manually, follow these steps:

1. Create a project directory and navigate into it:

BASHCopied
mkdir hello-mastra
cd hello-mastra

2. Initialize a TypeScript project with the required packages:

BASHCopied
# Using npm
npm init -y
npm install typescript tsx @types/node mastra@alpha --save-dev
npm install @mastra/core@alpha zod
npx tsc --init

Using the CLI

You can also use the mastra init command to create a new project:

BASHCopied
mastra init

This command can be run in three different ways:

1. Interactive Mode: Run without flags for a guided setup
2. Quick Start: mastra init --default for a setup with all components and OpenAI
3. Custom Setup: mastra init --dir src/mastra --components agents,tools --llm openai --example

Next Steps

After installation, you can start building your application with Mastra's components:

• Agents: For creating AI assistants with specific capabilities
• Tools: For extending your agents with custom functionalities
• Workflows: For orchestrating complex interactions between agents and tools

The framework is designed to work with various LLM providers, allowing you to choose the one that best fits your needs.