Mastra
: How to use it, features, and the business problems it solves

How to Use

1. Create a project

   Run the following command in your terminal to create a Mastra project.

   npx create-mastra@latest
   

2. Initial setup

   Interactively choose your project name, the components to use (Agents, Tools, Workflows, etc.), and your default LLM provider (OpenAI, Anthropic, Google Gemini, etc.).

3. Build an agent

   Define your agent in TypeScript and incorporate the tools and workflows you need.

4. Design workflows

   Build complex workflows that include conditional branching and loops to control your agent's behavior.

5. Deploy and operate

   Deploy your agent in a local environment or in the cloud and begin operating it.

Features

• TypeScript-based AI agent development

  Provides everything you need for AI agent development in an all-in-one package within a JavaScript/TypeScript environment, including building agents, integrating tools, information retrieval via RAG (Retrieval-Augmented Generation), and assembling workflows.

• Support for multiple LLM providers

  Supports multiple LLM providers such as OpenAI, Anthropic, and Google Gemini, letting you easily switch the backend AI model just by changing a code setting. Zenn+1파이토치 한국 사용자 모임+1

• Voice feature integration

  You can add voice input and output capabilities to your agents, enabling the development of applications with voice interfaces.

• Flexible workflow control

  Flexibly design and control complex workflows, including conditional branching, loops, retries on error, and waiting for human input.

• Storage and monitoring

  Provides a storage system to manage agent activity history and data, making evaluation and monitoring straightforward.

Pricing

Mastra is an open-source framework and is free to use.