OpenSpec is about to get 10x better.
We've spent the past year learning from developers and teams using OpenSpec every day.
Now we're bringing those lessons into our own custom-built agent harness.
Our goal is simple: help developers stay involved in the decisions that matter, better understand the work being done, and ultimately have greater confidence in what agents produce.
A completely new way to experience OpenSpec.
We'll be releasing an early preview next week. Stay tuned!
Absolutely don't get the "team of agents" trend. I've never wanted to manage a team or ever have an agent for different roles.
I never want to have to think, "Which agent should I send this message to?".
Having a single agent that can play multiple roles feels much more sane to me.
It's so much more powerful to have one agent be anything you want in the moment that can cross roles/boundaries effectively.
Tabish retweeted
6 months ago, we launched Rhizome AI to help everyone building a drug or device know exactly what their regulator wants them to do.
Since then, we've doubled countries covered, tripled the user base, and 5x'd ARR. But there's a lot more people we still want to help.
Announcing the Rhizome AI Free Plan: create an account and you get 10 questions/month.
Try it out: rhizomeai.com
Tabish retweeted
Very excited to publish the latest #UnhandledException podcast episode! I was honoured to have been joined by @0xTab, the creator of @OpenSpec_! A framework I use daily! It's amazing, so I really enjoyed geeking out with Tabish about it 😊
unhandledexceptionpodcast.co… #ai
Tabish retweeted
Enjoyed an awesome podcast discussion just now with @0xTab, the creator of @OpenSpec_! We covered A LOT! - really looking forward to publishing this one!! Should be out in about a week 😊🎙️ #UnhandledException #ai













