Back to Repositories

mastra-ai/mastra

Mastra is a TypeScript-first framework for building AI-powered applications and autonomous agents, created by the team behind Gatsby and backed by Y Combinator (W25 batch). With 21,800+ GitHub stars and 13,300+ commits, it has rapidly become the go-to choice for JavaScript and TypeScript developers who want to build production-ready AI agents without switching to Python. Mastra connects to over 40 LLM providers — including OpenAI, Anthropic, and Gemini — through a single unified interface, eliminating vendor lock-in. Its graph-based workflow engine lets you orchestrate multi-step AI pipelines with intuitive syntax, while built-in human-in-the-loop capabilities allow workflows to pause, persist state, and resume after human review. The framework supports the Model Context Protocol (MCP) for exposing tools and resources, integrates natively with React, Next.js, and Node.js, and can be deployed as a standalone server or bundled into existing apps. What sets Mastra apart from Python-centric alternatives like LangChain or CrewAI is its seamless fit into the modern TypeScript ecosystem — if you already build with Next.js, you can add AI agents to your stack without context-switching languages. With built-in evaluation and observability tooling, RAG support, conversation memory management, and support for Vercel AI SDK UI and CopilotKit, Mastra covers the full lifecycle from prototype to production.

TypeScript

Why It Matters

Mastra fills a critical gap in the AI agent ecosystem: a production-grade, TypeScript-native framework for the millions of JavaScript developers who want to build AI applications without adopting Python. Backed by the Gatsby team's deep React/Node expertise and Y Combinator, it is the fastest-growing TypeScript AI agent framework on GitHub, with 21.8k stars and 1.7k forks as of March 2026.

Repository Stats

Stars
22.0k
Forks
1.7k
Last Commit
3/14/2026

Related Resources

Weekly AI Digest