We just open-sourced the skill our team uses to draw diagrams.
It’s called draw-architecture 🙌
It turns a text description of your system architecture into a clean, interactive canvas diagram in one go:
– Drag, zoom, and click to isolate components — feels just like Figma
– One-click high-res PNG export, ready for slides or docs
– Best part: share the link with your team and everyone can leave comments right on the diagram
If you also draw architecture diagrams a lot, worth to give it a shot
I've been using @capydotai daily on all my GStack/GBrain PRs and I'm going at least 4x faster than I was when just using raw Codex/Claude Code via Conductor.
On a recent PR, this is what it did that out of box Codex wouldn't have done:
Now you can create videos the way you vibe-code.
Meet Pexo, your AI video agent.
Share your vision and references. Work through the details with Pexo like you’re messaging a friend.
No complex prompts. No new tools to learn.
Want to change something?
Mark it on the video and tell Pexo what to change, just like leaving a comment in a doc.
Everything in this video was made with Pexo:
the visuals, the motion graphics, the music, the captions, and the voiceover.
Even our founder @evanLiaoQ appears on screen.
You vibe-coded what you built.
Now vibe-create how the world sees it.
#Pexo #AIVideoAgent #VibeCreate #MadewithPexo
A bakery owner and an app founder hit the same wall with paid ads.
Neither has time to figure it out. Both need customers yesterday. 👇
https://t.co/vnx5jNTfcB
450K+ views. I didn’t touch the timeline once.
This entire launch video was generated in Diffusion Studio.
And the project is fully open source. Take it apart. Make it yours.
Zuck opening Muse connectors is the App Store for the AI era.
Soon you just say what you need on your phone, and the agent picks the product and checks out for you.
That brings up the merchant trust layer: will the agent pay, will the merchant accept the agent, and after something goes wrong, can anyone audit it independently.
All three are still tied to platform logs. Muse has Sentinel, one-time cards, and an in-app trail.
The next wave may sit right here 👇
Opening access for developers to build Muse connectors. You bring the API -- Muse brings the agent, the browser, and the context of what the person actually wants. People reach your service just by asking for it, and their agent takes it from there.
New connectors are live today. Come build with us. https://t.co/o6oTf1Sj9y
@finkd The next wave may sit right here👇
1. Agent SEO or GEO
2. Verifiable identity for merchants and agents
3. Audit
Check the details:
https://t.co/itKli3LZAd
Zuck opening Muse connectors is the App Store for the AI era.
Soon you just say what you need on your phone, and the agent picks the product and checks out for you.
That brings up the merchant trust layer: will the agent pay, will the merchant accept the agent, and after something goes wrong, can anyone audit it independently.
All three are still tied to platform logs. Muse has Sentinel, one-time cards, and an in-app trail.
The next wave may sit right here 👇
The next wave may sit right here👇
1. Agent SEO or GEO: who gets chosen
2. Verifiable identity for merchants and agents: is this shop real, and is this a real agent or a hostile bot
3. Audit: can a third party replay the user’s original intent, what the agent actually did, and the payment settlement (this is where blockchain-related protocols have a natural edge)
We're adding support for AGENTS.md to Claude Code.
Starting today in version 2.1.277, if there is no CLAUDE.md in a folder, Claude will check for and use AGENTS.md.
You can toggle this behavior in /config.
Breaking: Browser Use + Jev = Ultrafast ⚡
Findings flights took 7s and cost only $0.0039 🤯
> new action space every step
> DOM state space
> small LLM fallback to type
(this video is at 1x speed btw)
Built a tiny open source browser agent. try it below ↓
Any scene that needs to understand input and then make a pure judgment can be changed to it, which is extremely fast and cheap, with 0 output tokens.
Brilliant!!!
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution