# mastra-ai/mastra

Mastra is the modern TypeScript framework for AI-powered applications and agents.

Repository: https://github.com/mastra-ai/mastra
Canonical: https://ross.abutalabs.com/products/mastra
Homepage: https://mastra.ai
Language: TypeScript
License: NOASSERTION
License Family: other
Topics: agents, ai, chatbots, javascript, llm, nextjs, nodejs, reactjs, typescript, workflows, evals, mcp, tts
Last push: 2026-08-27T00:32:58+00:00

## Health v2 (maintenance only)
Score: 86/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 54
- inputs: {"age_days": 757, "days_push": 7, "days_rel": 7, "gap_med": 4.0, "n_releases_24m": 113}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 27495, forks 2697 (observed 2026-08-28T04:11:47.778481+00:00)

## What it is
Mastra is a TypeScript framework for building AI-powered applications and autonomous agents, with model routing across 40+ providers, graph-based workflows, memory, RAG, and MCP support. It includes observability, evals, and deployment tooling to take agents from prototype to production.

## Use cases
- build ai agents in typescript
- orchestrate multi-step llm workflows
- add rag and memory to a chatbot
- route llm calls across multiple model providers
- add human approval steps to agent workflows
- evaluate and monitor llm agent behavior
- build a chatbot with nextjs

## When to choose
- you're building AI agents or LLM apps in TypeScript with React/Next.js or Node
- you need durable workflows, memory, RAG, and evals in one framework
- you want provider-agnostic model routing across many LLM vendors
- you need human-in-the-loop suspension and resumable agent state

## When to avoid
- your stack is Python-first (LangChain, LlamaIndex may fit better)
- you need a simple single LLM call without agent orchestration overhead
- you require a permissive license for enterprise features, which are separately licensed
- you want a fully managed agent platform rather than a self-hosted framework

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, rag, workflow-automation, mcp, chatbot, web-framework, monitoring, testing
- domain: artificial-intelligence, large-language-models, chatbots, web-development, developer-tools
- platform: cross-platform, cloud, self-hosted
- tags: typescript, llm-orchestration, model-routing, memory, human-in-the-loop, evals, observability, nextjs, react, ai-agents, retrieval-augmented-generation, nodejs, web-server

## Member repositories
- mastra-ai/mastra (main) score 86

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:47.778481+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:54:31.781651+00:00, confidence not recorded.
  - readme: https://github.com/mastra-ai/mastra (fetched 2026-08-28T04:11:47.778481+00:00, sha 40914233cbed)
  - homepage: https://mastra.ai (fetched 2026-08-29T07:51:55.705821+00:00, sha d65f9d3d181e)
  - site_page: https://mastra.ai/docs (fetched 2026-08-29T07:51:55.718050+00:00, sha a85669b7d09b)
  - site_page: https://mastra.ai/guides/getting-started/quickstart (fetched 2026-08-29T07:51:55.720100+00:00, sha c273951df844)
  - site_page: https://mastra.ai/blog/category/changelogs (fetched 2026-08-29T07:51:55.722447+00:00, sha 908b375584c2)
  - site_page: https://mastra.ai/pricing (fetched 2026-08-29T07:51:55.715607+00:00, sha 69bae58fb0f2)
- Data as of 2026-08-30T08:39:29.467469+00:00.