by calcsam on 2/19/25, 3:25 PM with 154 comments
You can start a Mastra project with `npm create mastra` and create workflow graphs that can suspend/resume, build a RAG pipeline and write evals, give agents memory, create multi-agent workflows, and view it all in a local playground.
Previously, we built Gatsby, the open-source React web framework. Later, we worked on an AI-powered CRM but it felt like we were having to roll all the AI bits (agentic workflows, evals, RAG) ourselves. We also noticed our friends building AI applications suffering from long iteration cycles: they were getting stuck debugging prompts, figuring out why their agents called (or didn’t call) tools, and writing lots of custom memory retrieval logic.
At some point we just looked at each other and were like, why aren't we trying to make this part easier, and decided to work on Mastra.
Demo video: https://www.youtube.com/watch?v=8o_Ejbcw5s8
One thing we heard from folks is that seeing input/output of every step, of every run of every workflow, is very useful. So we took XState and built a workflow graph primitive on top with OTel tracing. We wrote the APIs to make control flow explicit: `.step()` for branching, `.then()` for chaining, and `.after()` for merging. We also added .`.suspend()/.resume()` for human-in-the-loop.
We abstracted the main RAG verbs like `.chunk()`, `embed()`, `.upsert(),’ `.query()`, and `rerank()` across document types and vector DBs. We shipped an eval runner with evals like completeness and relevance, plus the ability to write your own.
Then we read the MemGPT paper and implemented agent memory on top of AI SDK with a `lastMessages` key, `topK` retrieval, and a `messageRange` for surrounding context (think `grep -C`).
But we still weren’t sure whether our agents were behaving as expected, so we built a local dev playground that lets you curl agents/workflows, chat with agents, view evals and traces across runs, and iterate on prompts with an assistant. The playground uses a local storage layer powered by libsql (thanks Turso team!) and runs on localhost with `npm run dev` (no Docker).
Mastra agents originally ran inside a Next.js app. But we noticed that AI teams’ development was increasingly decoupled from the rest of their organization, so we built Mastra so that you can also run it as a standalone endpoint or service.
Some things people have been building so far: one user automates support for an iOS app he owns with tens of thousands of paying users. Another bundled Mastra inside an Electron app that ingests aerospace PDFs and outputs CAD diagrams. Another is building WhatsApp bots that let you chat with objects like your house.
We did (for now) adopt an Elastic v2 license. The agent space is pretty new, and we wanted to let users do whatever they want with Mastra but prevent, eg, AWS from grabbing it.
If you want to get started: - On npm: npm create mastra@latest - Github repo: https://github.com/mastra-ai/mastra - Demo video: https://www.youtube.com/watch?v=8o_Ejbcw5s8 - Our website homepage: https://mastra.ai (includes some nice diagrams and code samples on agents, RAG, and links to examples) - And our docs: https://mastra.ai/docs
Excited to share Mastra with everyone here – let us know what you think!
by Palmik on 2/20/25, 5:47 AM
testWorkflow
.step(llm)
.then(decider)
.then(agentOne)
.then(workflow)
.after(decider)
.then(agentTwo)
.then(workflow)
.commit();
by kylemathews on 2/19/25, 6:46 PM
Also the team is top-notch — Sam was my co-founder at Gatsby and I worked closely with Shane and Abhi and I have a ton of confidence in their product & engineering abilities.
by joshstrange on 2/19/25, 6:55 PM
by alanwells on 2/19/25, 8:32 PM
by brap on 2/19/25, 7:19 PM
Can anyone please give me a usecase, that couldn’t be solved with a single API call to a modern LLM (capable of multi-step planning/reasoning) and a proper prompt?
Or is this really just about building the prompt, and giving the LLM closer guidance by splitting into multiple calls?
I’m specifically not asking about function calling.
by Gakho on 2/19/25, 5:56 PM
by epolanski on 2/19/25, 6:59 PM
by _pdp_ on 2/19/25, 10:13 PM
This is like 5% of the work. The developer needs to fill the other 95% which involves a lot more things that are strictly outside of scope of the framework.
by harliem on 2/19/25, 6:01 PM
by eliotthehacker on 2/19/25, 8:49 PM
by aranibatta on 2/19/25, 3:58 PM
by netcraft on 2/19/25, 8:03 PM
by monideas on 2/19/25, 7:30 PM
by davedx on 2/19/25, 7:39 PM
by PetrBrzyBrzek on 2/20/25, 7:29 AM
by lmrl on 2/20/25, 7:09 AM
2. Related to previous question, since this is node based, is it possible to support websockets?
by orliesaurus on 2/20/25, 5:02 AM
by dstroot on 2/20/25, 7:39 PM
by dhorthy on 2/19/25, 8:57 PM
by _1 on 2/19/25, 6:35 PM
by realmikebernico on 2/19/25, 5:57 PM
by 5Qn8mNbc2FNCiVV on 2/19/25, 5:58 PM
by fnikacevic on 2/19/25, 10:20 PM
by cshimmin on 2/20/25, 7:45 AM
by tobyhinloopen on 2/20/25, 7:51 AM
by dikaio on 2/20/25, 5:42 PM
by levensti on 2/19/25, 6:22 PM
by asati on 2/20/25, 12:19 AM
by albertmz on 2/22/25, 1:05 AM
by gregpr07 on 2/19/25, 7:03 PM
by fuddle on 2/19/25, 7:20 PM
by delduca on 2/19/25, 11:14 PM
It started off wrong.
by jobryan on 2/19/25, 5:49 PM
by pablodecm on 2/19/25, 8:08 PM
by yovboy on 2/19/25, 5:47 PM
by bobremeika on 2/19/25, 5:40 PM