Gandy Marroquin S.

[Chema Alonso @chemaalonso]

El lado del mal - Naked Sun: Fragmented Jailbreak Attacks in Multi-Agents & Jailbreak Monitor (JAMON) https://t.co/QeQCkl0tdJ #FragmentedJailbreak #AgenticAI #MAS #Hacking #Hardening #IA #AI #LLM #Jailbreak

[Windows Developer @windowsdev]

More open-source model choice is coming to Windows. 💻

We’re bringing llama.cpp support to Windows ML, giving developers a faster way to experiment with emerging models and deploy across GPU, NPU + CPU.

Learn more: https://t.co/Kjuah6AI7f

[GitHub @github]

Devs and agents are moving faster than ever. Secret protection needs to keep pace.

GitHub’s new context-aware classifier checks candidate secrets in under 2 milliseconds and could more than double the number of secrets push protection prevents before they enter repository history.
https://t.co/7LeAg7nnAm

[Digitalearn @DIGITALEARN_1]

Cybersecurity Masterclass — from Zero to Advanced.

Build real-world skills in Security, Cloud, GRC & SOC

Download here: https://t.co/9FZTEhMkk5

[GitHub Projects Community @GithubProjects]

Run a complete email server in one container, configured entirely with files.

- Handles sending, receiving, filtering, and spam.
- No database; settings stay versioned and simple.
- Deploy and upgrade easily.

[Miguel Ángel Durán @midudev]

OpenAI ha publicado un curso gratis de Códex.
Te explica cómo desarrollar e iterar webs y apps.

✓ 22 lecciones. 3,5 horas. Sin tarjeta
✓ Skills, plugins y feedback visual para iterar menos

En inglés pero con subtítulos:
→ https://t.co/C7FoDBMzkw

[Google Analytics @googleanalytics]

What if answering your toughest performance questions simply required asking “why?” ⚡

With Ask Advisor in Google Analytics, powerful and detailed analysis is a prompt away, helping you skip the manual deep-dive and make smarter optimization decisions faster.

Learn more about how Ask Advisor can streamline your reporting → https://t.co/56LJGkrqf8

[freeCodeCamp.org @freeCodeCamp]

LLMs are pretty impressive tools. But on their own, they can't "remember" anything.

That's where vector stores come in and enhance their capabilities.

In this article, @manishmshiva explains what vector stores are and how they work, teaches you about embeddings, and more.

https://t.co/lnwmZCk4NP

[Chema Alonso @chemaalonso]

El lado del mal - MyPublicInbox: Sexto Aniversario https://t.co/3MLTl2SDkZ #MyPublicInbox #email #Correo #Comunicación @mypublicinbox

[Oracle Developers @OracleDevs]

Vector search can find relevant facts, but some questions also require proof of how those facts connect.

Explore when Graph RAG helps teams trace relationships across services, customers, and policies to deliver verifiable AI answers (via @thenewstack). https://t.co/fPsJyFUF3m

[Brais Moure @MoureDev]

¿Quieres diseñar diagramas profesionales desde tu agente de IA?

Diagram Design te permite crear 42 tipos diferentes con HTML + SVG utilizando Claude Code, Codex, Copilot, Pi... Todo manteniendo los contenidos uniformes y sin distorsión.

→ https://t.co/jcHqie4mLa

[El Pingüino De Mario @PinguinoDeMario]

El riesgo de no proteger correctamente los endpoints de nuestra api, viendo también como corregir este tipo de vulnerabilidades. #ciberseguridad #apirest #api #kalilinux

[Crimson Frontier Universe | @CrimsonFrontier]

Una decisión puede cambiar una amistad — y también a la persona en la que Adam se convierte.
Añade Crimson Frontier: Youth a tu lista de deseados en Steam.

[Windows Central @WindowsCentral]

The Windows App now has a new way to connect to your PCs!

Remote PC connections have arrived after Microsoft retired the Remote Desktop app, and the expanded feature set originally began testing in March. Now, it's available to anyone.

https://t.co/WzghrSeVcK

[GitHub Projects Community @GithubProjects]

Review AI coding work in your browser and send feedback straight back to the agent.

- Annotate plans and specs before coding
- Comment on code changes and pull requests
- Works with many AI coding agents

Explore it here:
https://t.co/4ERo3zb2MV

[freeCodeCamp.org @freeCodeCamp]

Getting an email sent successfully doesn't mean it'll actually reach an inbox.

In this article, @reetain_raina explains how AI is changing email filtering, sender reputation, and inbox placement.

You'll learn how SPF, DKIM, DMARC, sending behavior, and provider-specific signals affect whether your emails end up in the inbox or spam.

https://t.co/Hl3dFmKNfM

[Oracle Developers @OracleDevs]

Want agents that get smarter without rebuilding context every time? Treat memory as enterprise infrastructure. Give agents durable, shared knowledge with identity, access controls, governance, and lifecycle management built in.

Build once, reuse across agents.

This blog shares more details, including where to find our booth at AI Engineer New York! https://t.co/mWu0giqmFb

[Windows Developer @windowsdev]

The next era of the PC is taking shape on Windows. ✨
Windows is becoming the home for hybrid intelligence.
🔐 MXC GA
🧠 Frontier coding models running locally
✨ GitHub Copilot HydraFusion
💻 llama.cpp + Windows ML
⚡ Surface Laptop Ultra + Dev Box preorders
🖥️ New RTX Spark PCs

[Miguel Ángel Durán @midudev]

¡Nueva alternativa a Parallels! Crea máquinas virtuales de Windows y Linux en tu Mac.

✓ Optimizado para Apple Silicon y su GPU
✓ MCP en un clic (Claude Code, Codex, Gemini...)
✓ Crea snapshots automáticamente, para volver atrás

→ https://t.co/bWx1Aj19Ex

[Brais Moure @MoureDev]

¿Qué permisos debería tener un agente de IA para trabajar con tus archivos, tus APIs y tus credenciales?

NVIDIA quiere solucionar eso con OpenShell, un entorno de ejecución de código abierto que permite ejecutar agentes autónomos bajo políticas de acceso definidas.

Claves importantes:
- Aislamiento: cada agente trabaja en un entorno separado, con límites sobre archivos, procesos y conexiones de red.
- Protección de credenciales: el agente no recibe los secretos reales, ya que OpenShell los incorpora a las peticiones dirigidas a destinos autorizados.
- Control de permisos: utiliza verificación para detectar accesos potencialmente peligrosos antes de aplicar cambios implortantes.
- Integración: ofrece un CLI y SDKs para Python, TypeScript, Go y Rust.

La idea es dar a los agentes capacidad para actuar y establecer límites sobre lo que pueden hacer.

→ https://t.co/uXp9JqlbOM