🚨 @Karpathy predicted the power of the "LLM Wiki." Google just formalized it.
Meet Open Knowledge Format (OKF): a vendor-neutral standard for giving foundation models the curated context they need.
I can genuinely see this replacing Notion, Obsidian, or traditional wikis for developer teams, and the reason comes down to bookkeeping.
Traditional wikis fail because humans inevitably abandon the tedious work of updating them.
As Andrej Karpathy pointed out recently, LLMs don't get bored.
They don't forget to update a cross-reference, and they can touch 15 files in a single pass.
OKF standardizes the interoperability layer so agents can actually do that heavy lifting autonomously.
Because the format is minimally opinionated, it doesn't dictate what you write, it just dictates how it's structured. You get:
→ Human-readable documents that live right alongside your code in version control
→ Cross-links that map out complex entity relationships without needing a graph database
→ A system that survives moving between different tools and organizations
There is no complex compression scheme.
No central registry.
If you can cat a file, you can read it.
If you can git clone a repo, you can deploy it.
This is how we stop rebuilding context pipelines from scratch every time a new model drops.
Announcement + spec file in 🧵↓
Trying out my /teach skill today, imagining I was a vibe coder wanting to learn the basics.
Here are the four lessons it created so far:
1. It interrogated me on my mission - the reason why I wanted to learn vibe coding. I said I was a teacher wanting to build a scheduling app.
2. Started me off with git so I had the ability to roll back bad work. Just 5 basic commands. No rebasing/merging yet. Perfect.
3. Taught me the basic units of a full-stack app: frontend, backend, auth and database. Every example was tied to the app I wanted to build. Quiz at the end.
4. I didn't choose a stack preference, but it noticed Node was already installed! Clever agent. So it chose Next.js and Supabase as my stack, and taught me hot reload. I got commands to set up my app.
Each day sent me off to primary sources to verify that what I was learning was true. Git docs, AWS articles, Next.js docs.
This is addictive, personalized, and infinite. Give it a go.
Mi lectura de 'Enviado especial' @perezreverte@AlfaguaraES tiene sesgo.
Me resulta imposible no leer al escritor en primerísima línea.
La novela está hecha de eso que repugna, mancha y lastima. Eso que él vio y vivió en la guerra.
https://t.co/RUipvy4YVp
📷
Auto-memory in Claude and ChatGPT is the wrong default.
The model decides what to keep. You cannot see it, edit it, or take it with you.
A folder on your own disk does the job, in plain text, portable forever.
Walkthrough: https://t.co/nkoZ81sRvO
"Won't all those rules and guardrails slow the AI down?"
No. Structure is what makes speed possible.
My AI agents produce more in a day than most teams do in a week. BECAUSE of the guardrails, not despite them.
Gracias a @LaVanguardia por esta fotografía de conjunto. Es dolorosa, sin duda. Cómo no.
Cinco novelas para entender la Venezuela actual
https://t.co/JS6AmhbKk7
En efecto, lo mismo que a los ojos de los murciélagos ciega la luz del día, lo mismo a la inteligencia de nuestra alma ciegan las cosas que tienen en sí mismas la más brillante evidencia.
So far private credit has mostly consisted of private-equity firms lending to companies owned by other private-equity firms. But now they view the $13trn of loans sitting on the balance-sheets of banks as fair game https://t.co/WAI3Vo9U8O
Hoy mi sesión de noche, después de llevar todo el día dando la turra con Gary Oldman, tenía que ser una de él👇🏼 ¿Les gusta?♥️🙌
Mil gracias por todo el cariño, de verdad. Son ustedes inmensos🫂🫂♥️♥️ Feliz noche😊