Introducing @𝚓𝚜𝚘𝚗-𝚛𝚎𝚗𝚍𝚎𝚛/𝚌𝚑𝚊𝚝
Generative UI for Slack, Teams and more
→ Agents reply with native cards
→ Tables, fields, charts and buttons
→ Streams in as the model writes
→ One catalog for every Chat SDK platform
Prompt of the day is a simple one:
"How do we make this change fully verifiable?"
Makes the agent bend over backwards to verify all its changes. Signing up for real services, clicking through the frontend. Quality goes through the roof.
🚨this is going to END some careers.
a dev just dropped a free Claude skill called "teardown" and anyone who animates 3d for money should pay attention.
the skill is ridiculously good but the dev is crazy enough to give it away anyway.
how it runs:
> you hand it a picture and say what the creature is
> claude codes the octopus, renders a still beside your picture and patches the differences round after round before any animation
> every arm is a chain of links that shrink toward the tip, so each one curls into a true spiral with 2 rows of suction wheels underneath
> one timeline drives the whole thing: the jet gauge, the operator's cursor, the console rows and the audio all tick off it
> it can't watch footage, so it reviews a grid of frames at every beat, fixes what looks off and re-renders until a full pass comes back clean
> the frames come off your own GPU at 60 fps and python writes the soundtrack on the same beats
one picture went in and this octopus is what came back out.
free skill, full write-up in the article 👇
We built Pando, a new Git backend for our monorepo, to handle more pushes and fetches without changing how our engineers and agents use Git.
Pando cut push latency by ~80% and now serves nearly a petabyte of Git data a day. We can grow Git capacity alongside agent usage.
@alejandrozepe Hay algunas apps como YNAB que se pagan solas con la organización que te enseñan a tener.
Migré hace poco a SURE selfhosted y luego cuando no tengo red o luz extraño el servicio en la nube jaja
We’re adding support for MCP Tasks and Elicitations in Claude.
Tasks let Claude give plugins long-running jobs such as deep research and report back when they’re done.
Speaking of security, we created a registry for building OIDC configuration components.
Pick your providers, form and validation libraries, and token verifier (powered by jose). Then run one command to copy everything into your app.
I've built these a few times before. I know how annoying they can be to get right. Hope this helps.
https://t.co/pLIq4PjKPO
Let me put my TypeScript hat back on for a minute
Content mappers look outrageously exciting. This is now possible:
1. import any non-TypeScript file (markdown, vue, svelte)
2. Write a content mapper to extract types from it
3. Run tsc with --runExternalCode
4. You get type inference from NON TYPESCRIPT CODE
The simplest case:
1. Import a markdown file
2. Add a bit of config to tsconfig.json
3. Run tsc with --runExternalCode
4. Get fully typed markdown frontmatter
Unbelievably useful, a dreamland for tinkerers
Opus 5.5 + /anatomy skill is insane for motion animation
• Interactive isometric wireframes you can rotate, inspect and explore
• WebGL shaders for fire, water, plasma and more
• Real 3D motion, with individual parts that rotate
send to Claude:
https://t.co/3c9pJq05FV
Such a good skill for visual explainers. All SVG and animated.
Prompt I used:
https://t.co/CvOESSiJFY install skill
/anatomy Explain how Seedream 5.0 is able to split an image into layers and what cool things you can produce with it. Do that in a single HTML file.
Change the camera/zoom in for scene 1 to 5 to explain more clearly.
I still spend a LOT of time figuring out the right abstractions and interfaces (user and programming).
I recently built two components for shadcn/react. Each took weeks of UX research, experimentation, and iteration to get right.
AI hasn't accelerated this part yet, at least for me.
What it has done though, is help me fail faster. I can now go through many more failures before a component or code is ready.