
Ever tried. Ever failed. No matter. Try again. Fail again. Fail better. - Samuel Backett

making things with AI; cofounder & ceo @onext_ai; cofounder & ceo beautifulflywheel.com

Ever tried. Ever failed. No matter. Try again. Fail again. Fail better. - Samuel Backett

Introducing eve 1.0 Build production agents.
eve is generally available with v1.0. • Build agents on your own harness • Ship multi-tenant, multiplayer agents • Govern & trace sessions with Agent Runs • Render agent UIs via Next.js, Nuxt & more Read the blog → vercel.com/blog/eve-is-now-g…

De 9.791 hallazgos a 99 informes. El trabajo caro siempre fue el triage, y un detector más ruidoso no lo abarata
Maintainers need fixes, not more AI bug reports. With @msftsecurity, we turned 9,791 raw AI security findings into 99 verified, private reports with suggested fixes for six open source projects. And maintainers are already shipping fixes. What we learned from Codename MDASH, a new security project 🔍 github.blog/open-source/main…

Un script de tests A/B corre amb els mateixos permisos que el codi de la teva web. 16 hores servint malware a 178 empreses sense tocar-los els servidors
Ikea, Zoom, la BBC i altres empreses distribueixen programari maliciós durant hores 3cat.cat/3catinfo/ikea-zoom-… 3cat.cat/3catinfo/ikea-zoom-…

Encadenar herramientas en un script en vez de llamada a llamada ahorra muchas idas y vueltas por el contexto. La pregunta es con qué permisos corre ese script
Check out Code Mode, a new, experimental way for Claude Code to work with tools. Claude writes scripts that chain your tools together in sophisticated ways, and it can answer you while long commands are still running. Opt in with CLAUDE_CODE_EXPERIMENTAL_CODE_MODE=1 code.claude.com/docs/en/code…

The highly classified network that the National Security Agency uses for its internal communications is down, according to six people familiar with the matter. At this point there is no evidence of a cyberattack, sources said. CNN has asked the National Security Agency for comment. cnn.it/4AA0eTd

We're teaming up to accelerate open-source Al. OpenRouter is now integrated into Hugging Face directly. Same token, same endpoint, plus live model stats on the Hub and an inference widget we redesigned together. The best of open, in one place. huggingface.co/blog/openrout…

Modelo y harness del mismo fabricante, y europeo. El benchmark que me importa es cuántas reglas del repo respeta sin recordárselo
To celebrate Mistral Vibe + Le Chonk becoming the world’s best open model-harness duo, we’re making Vibe free in Europe for the first 3 months. (Yes, it’s actually better than GLM / DeepSeek at coding). Use code LECHONK via link in 🧵. Limited time, first-come, first-served.

Quedan pocas plazas para el taller online de SDD del jueves 26 (16:30-18:00, Teams, gratis). 90 minutos para ver cómo un equipo de desarrollo trabaja con agentes a partir de especificaciones, de la idea al código revisado. Si te interesa, apúntate ya onext.es/es/taller-sdd-onlin…

Lo mejor está al final. Misma nota de calidad y más llamadas a herramientas con parámetros inválidos. La media de las evals lo escondía, la traza no
Datadog Brings Production Experimentation to AI Agent Observability bit.ly/4iULzL8 by @RisiCraig #aiops

Agent skills let you capture repeatable instructions so you and your agents can reuse them. A practical guide on agent skills to keep past instructions, institutional knowledge, and other context 👇


Muy buen artículo. Que el código del agente se ejecute fuera de tu portátil debería ser lo normal
I've been exploring new ways to sandbox code written by coding agents - here's a description of the external sandbox pattern, where agents run on your Mac but execute code in a WebAssembly or QEMU or Apple Container sandbox - with demos for Claude Code, Codex and Pi blog.primeradiant.com/extern…

We’re working with @ApolloResearch, @METR_Evals, @TransluceAI, and @GoodfireAI on embedded assessments: giving outside experts deep access to our internal models and systems to investigate alignment and test our safeguards. These assessments build on our ongoing alignment and safety research, bringing independent scrutiny to how we identify and manage risks in our systems. openai.com/index/our-first-p…

Tenemos 70 skills en onextOS y el validador solo comprueba que estén bien escritas. Si funcionan bien, hoy lo sabemos revisando a mano
NVIDIA ha lanzando su propia herramienta para comprobar la calidad de las skills de agentes y medir sus resultados. → github.com/NVIDIA/SkillEvalu…

El gran examen de Softcatalà per trobar la millor IA en català ✍️ @BernatCastanyer naciodigital.cat/societat/el…

🔒 Check your repo's code scanning alerts without leaving the GitHub Copilot app. With the new GitHub security extension, you can ask Copilot to find, explain, and fix (or dismiss!) an alert. Now in public preview ⬇️ docs.github.com/copilot/how-…