
Security News
Lovable’s OJ Rewrites Vite’s Dev Server in Rust as AI Lowers the Cost of Forking Open Source
Lovable’s OJ rewrites Vite’s dev server in Rust, reducing memory use and preview times as AI lowers the cost of open source reimplementation.
The project brain for AI-driven development: a brief becomes approved analysis documents, then your own agent CLI (Claude Code, Codex, Gemini) writes the code.
Building products with AI, I kept hitting the same two walls: I could not see where the project stood, and at some point the model would drop the rules we had agreed on and start doing its own thing.
I know you have the same problem. You just have not named it yet. I have. Its name is Kortext.
Kortext is a product constitution you agree on with the AI before development starts.
From the BRIEF.md you provide, 10 expert personas write 14 architecture documents — the one source of truth the model comes back to, every time.
If the brief you wrote is not enough, Kortext says so. The panel explains BRIEF.md and carries an example.
There is one human role, Prime.
The other personas: product manager, architect, designer, growth expert, security engineer, DevOps engineer, DBA, compliance expert, copywriter and QA engineer.
From the brief, Kortext produces PRODUCT.md, STACK.md, STRUCTURE.md, ARCHITECTURE.md, DESIGN.md, GROWTH.md, SECURITY.md, ENVIRONMENT.md, DATABASE.md, API.md, LEGAL.md, CONTENT.md, ENGINEERING.md, TEST.md. And, at the very end, an optional EXPERIENCE.md.
Every document written is put in front of you to review. Do not skip them — read. Trust me, it is worth it.
A document may carry questions addressed to you, a change request for a document written earlier, or a note for one still to be written. Those are the model's own findings. Beside them, you can select any line, ask a question and ask for a change. If the model asks for a change in a document and you accept it, that document goes back through the pipeline. Because the documents depend on one another, a document sometimes comes back to you more than once, like a loop. That spends a fair number of tokens. But it spends them before the hours and tokens that would otherwise be wasted later. On the whole, a much better deal.
When the writing is done, your project folder holds a .kortext directory and, inside it, your product constitution.
From there on you carry on building with whatever AI tool, in whatever client, you already use.
At that point Kortext's job is done. No sign-in, no key, no API. Kortext collects nothing and keeps no telemetry. I am very curious how much it gets used, though — and if that ever happens, it will be anonymous, no need to worry.
And I have not forgotten projects that are already under way. Kortext can start from an existing codebase too, and may come back with suggestions on where the project stands.
I build Kortext alone, in the evenings, and ship it under MIT. If it turned out to be the right guide for your project, that is exactly what I was after. May your products be great.
MIT © Eray Endes
FAQs
The project brain for AI-driven development: a brief becomes approved analysis documents, then your own agent CLI (Claude Code, Codex, Gemini) writes the code.
The npm package kortext receives a total of 1,056 weekly downloads. As such, kortext popularity was classified as popular.
We found that kortext demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.

Security News
Lovable’s OJ rewrites Vite’s dev server in Rust, reducing memory use and preview times as AI lowers the cost of open source reimplementation.

Security News
It has been one year since Shai-Hulud made its first appearance on npm.

Research
/Security News
Operators behind PolinRider used a compromised GitHub account to plant malware in four development versions of a Packagist package with 700,000+ downloads.