Would you believe me if I told you that this game was written in almost 100% pure Zsh?
There is only one extra module, zdraw, that I created and it enables really awesome graphics in a terminal.
https://t.co/kSo7DEzkI6
#zsh#games#vibecoding
This project started out because I wanted to see what Zsh was capable of and it turns out that it's really capable.
zcoder.zsh works and it's especially good for sysadmin stuff, which is where I use it the most.
This thing happened because I wanted to see how far I could push Zsh and just to have fun doing it.
Say hello to zcoder.zsh. A complete agent harness written in Zsh with a minimal set of external deps.
Try it: https://t.co/Pa6TyCJFI0
#zsh@ollama#localai#ai#agentharness #harness
It supports tool calls, patches, MCP, color coded code snippets and more. And it's incredible fast and resource efficient with only a handful of external deps.
Tested in Linux, but should work on Mac as well.
It's interesting how a silly thought turned into something incredibly useful. When I created zcoder.zsh I didn't imagine that it would be able to do what it does.
https://t.co/GvpY80M4n2
An AI coding harness written in Zsh.
Today I’m launching Atlas Scout - a local, read-only MCP server that gives coding agents a structural map of your codebase.
Less blind grep. Fewer wrong files. Better navigation.
Free tier + 14-day Pro trial:
https://t.co/UBwboAofqu
#MCP#AtlasScout#Release
Most AI security audits are broken.
They flag everything as a vulnerability and miss what actually matters.
We built a security audit skill that forces models to reason about trust boundaries and exploit value.
https://t.co/agvY2QTVgN
#AISecurity#LLMSecurity#AppSec
Most AI security audits are broken.
They flag everything as a vulnerability and miss what actually matters.
We built a security audit skill that forces models to reason about trust boundaries and exploit value — not just scary APIs.
https://t.co/W6ayQAfbFk
#AISecurity #LLMSecurity #AppSec
Give Zaguán Blade v0.5.0 a spin.
Its on its way to become what o set out to achieve when I started out.
https://t.co/MxwiWpP1oB
https://t.co/ufQTjaYZrK
I'm pleased to announce the release of Zaguán Blade v0.5.0.
A lot of work has gone into this release. It launches faster, uses less resources and contains a symbols search.
https://t.co/pNdUPVWfsx
https://t.co/MPRhrklv3g
I'm pleased to announce the release of Zaguán Blade v0.4.0.
Notable changes are support for themes with 4 themes built in. In addition to this there are also a lot of bug fixes and enhancements.
https://t.co/D8SlJwCNuC
#zblade#vibecoding#notvscode#AI#ZaguanBlade #NewRelease
Been an interesting few days working on Zaguán Blade and introducing so many cool new models.
Hope to see you enjoy and create with it.
https://t.co/JgqqPModbm
Mercury 2 is live.
The world's first reasoning diffusion LLM – 5x faster than leading speed-optimized autoregressive models.
Built for production: multi-step agents without delays, voice AI with tight latency budgets, instant coding feedback.
Diffusion-based generation enables parallel refinement, not sequential tokens. Faster. More controllable. Dramatically lower inference cost.
Available today on the Inception API.
@dinabass has the story in @business.
I'm proud to announce the very first alpha release of Zaguán Blade. It's my attempt at creating an AI Editor that is not based on VSCode.
Blade is written in Rust and uses Tauri for the frontend.
Yes, there are many bugs but it works.
Check out https://t.co/d5I6fLPUBq
#zblade
Been working on this awesome little thing for a little while and released v1.0.0 yesterday.
It's called tstlai - TypeScript Translator AI - and hooks into your existing code with only a few lines of code.
Check it out at npmjs.
MIT license.
I did a deep code audit of my main server today and was able to remove over 2700 lines of dead code, refactor a lot of common functionality and in make the whole codebase more maintainable.
Felt good to finally get it done.
The updated server in now in production.
#ZaguanAI#AI
I'm pleased to announce tstlai v1.0.0. It's working beyond my imagination and has become so much more than I initially set out to do.
https://t.co/5GJ3nbJbgr
#ZaguanAI#ZaguanLabs#tstlai#translations#AI
No more tens of JSON files for each language you want to support? ✅
Easy integration in your existing code? ✅
Live translations or save for offline use? ✅
Check out this video of tstlai in action.
#ZaguanLabs#AI#tstlai@Zaguan_AI@theo@OfficialLoganK
Looking forward to the meet up in Málaga on December 2nd with @_ai_collective and @windsurf. Never been to an event here in Spain and am really excited to meet new people, perhaps new friends, exchange ideas and share what I've been doing lately.