Patrick Vilchez

Helpdesk TI 💻 - DBA

Aliancista de Corazón 💙💙

[Gentleman Programming @G_Programming]

Regalito de finde :
https://t.co/eYZN41qIpE

Vi que mucha gente tenía dudas de las buenas prácticas que aplicamos en Gentle Shell a la hora de la utilización de sub agentes y cómo orgánicamente hacemos que esta metodología aplique de forma correcta, solo cuando se necesita.

Por ello les regalo esta guía documental para entender cómo funciona Gentle Shell a la hora de delegar y utilizar sub agentes + recomendaciones en cuanto a qué modelos elegir para cada uno.

Espero que les guste!!!

[Expe @Expefutbol]

Siguen vivos para el navegador:
BO1 Zombies: https://t.co/7BEvh1gS6Q
Modern Warfare 2: https://t.co/QuMBjfBU6b
Skate 3: https://t.co/KbrFMkph1k
CS Surf: https://t.co/6hnaUaVYVo
PES 6: https://t.co/vnGP8Be4F2
GTA 5: https://t.co/3BxdOnza24
GTA Vice City: https://t.co/O32MwiS8O5
GTA 3: https://t.co/vpeUVDVwt4
The Simpsons: Hit & Run: https://t.co/oatgihRzuG
Quake 1: https://t.co/Eb9dTIpc6N
Quake 2: https://t.co/054utnhfBW
Quake 3: https://t.co/ailUibeU4I
Return to Castle Wolfenstein: https://t.co/DZvzwvYYTP
Unreal Tournament: https://t.co/qsOXNihqKr
Half Life: https://t.co/J9rcnNWgdQ
Half Life / CS 1.6: https://t.co/yP3RJDfKo4
Diablo: https://t.co/OXdIilK3E6
Hedgewars: https://t.co/zIZdShy6ix
Among Us: https://t.co/y0MjGX9SLQ
Tomb Raider (OpenLara): https://t.co/w452TJct3c
Minecraft: https://t.co/NyBUs883MK

[Patrick Vilchez @PvilchezV]

@ColectivoWORLD @Universitario @raulchavezpa tus dtsb ganaron, igual los botaron jajaja

[Patrick Vilchez @PvilchezV]

@MuniLima por que , las armys no limpian tambien?

[Miguel Ángel Durán @midudev]

Todos los productos de Adobe reimplementados desde cero, gratuitos y de código abierto

→ https://t.co/p6yYSQtIm5

[Patrick Vilchez @PvilchezV]

@FaztTech @grok por ahora windows por instalaciones y aplicaciones gubernamentales

[Gentleman Programming @G_Programming]

Que es gentle-ai ?

Hecho por alguien de la comunidad ❤️‍🔥

[Patrick Vilchez @PvilchezV]

@ErrorArbitralTR cuanto te han pagado?

[Gentleman Programming @G_Programming]

@Ubendev You have gentle ai which works everywhere and Gentle Shell, our own custom agent:

https://t.co/kKINXa2nns

https://t.co/OaMIC9YeXH

Give it a try !

[Miguel Ángel Durán @midudev]

¡Xiaomi suelta MiMo-V2.6 y el precio es espectacular!
Pro pelea con GPT-5.6, Opus 5 y el nuevo Grok 4.7

Pro: 0,44 $ / 0,87 $
Flash: 0,14 $ / 0,28 $

✓ Flash gratis en OpenCode Zen
✓ Open source de verdad: pesos, paper y código de RL

→ https://t.co/fLChg89Yuh

[Patrick Vilchez @PvilchezV]

@midudev si no te pones al dia la tegnologia te djea

[Patrick Vilchez @PvilchezV]

@ntrujillo_pe se cae a pedazos el grupo

[Patrick Vilchez @PvilchezV]

@FaztTech bloc de notas

[Miguel Ángel Durán @midudev]

He recopilado 179 libros gratis de programación en español.

PDF, HTML y EPUB. 42 temas: JavaScript, Python, TypeScript, React, Rust, Git, SQL…

Sin registro, sin pagos, sin sorpresas:
→ https://t.co/OhHoYuehXp

[Gentleman Programming @G_Programming]

ODD disponible en gentle-ai y Gentle-Shell en la rama main !!

A probar se ha dicho

https://t.co/kKINXa2nns

https://t.co/OaMIC9YeXH

[Python Coding @clcoding]

Download YouTube videos using Python

[Gentleman Programming @G_Programming]

Esta es mi actual config usando los modelos de OpenAi que me funciona muy bien:

Orquestador: model=openai-codex/gpt-5.6-sol, effort=medium
sdd-init: model=openai-codex/gpt-5.6-sol, effort=medium
sdd-onboard: model=openai-codex/gpt-5.6-sol, effort=medium
sdd-explore: model=openai-codex/gpt-5.6-terra, effort=high
sdd-research: model=inherit, effort=inherit
sdd-proposal: model=openai-codex/gpt-6-astra, effort=low
sdd-spec: model=openai-codex/gpt-5.6-terra, effort=high
sdd-design: model=openai-codex/gpt-6-astra, effort=low
sdd-tasks: model=openai-codex/gpt-5.6-terra, effort=high
sdd-status: model=openai-codex/gpt-5.6-sol, effort=medium
sdd-apply: model=openai-codex/gpt-5.6-terra, effort=high
sdd-verify: model=openai-codex/gpt-6-astra, effort=low
sdd-sync: model=openai-codex/gpt-5.6-sol, effort=medium
sdd-archive: model=openai-codex/gpt-5.6-luna, effort=high
jd-judge-a: model=openai-codex/gpt-6-astra, effort=low
jd-judge-b: model=openai-codex/gpt-6-astra, effort=low
jd-fix-agent: model=openai-codex/gpt-5.6-terra, effort=high
gentle-ai-explore: model=openai-codex/gpt-5.6-terra, effort=high
gentle-ai-verify: model=openai-codex/gpt-6-astra, effort=low
gentle-ai-worker: model=openai-codex/gpt-5.6-terra, effort=high
review-readability: model=openai-codex/gpt-6-astra, effort=low
review-refuter: model=openai-codex/gpt-6-astra, effort=low
review-reliability: model=openai-codex/gpt-6-astra, effort=low
review-resilience: model=openai-codex/gpt-6-astra, effort=low
review-risk: model=openai-codex/gpt-6-astra, effort=low
review-validator: model=openai-codex/gpt-6-astra, effort=low

[Gentleman Programming @G_Programming]

You just described exactly what we've been building for months, so let me be the annoying guy who says "we have that".

It's called Gentle AI. It's the harness, not the model. A Go core that holds the discipline so you don't have to: you say "I want this" and it takes you there with every best practice applied, and you never need to know they exist.

It also has a memory layer, super fast, robust and efficient, called "Engram" https://t.co/0dgeRTQ47V

The part nobody else has is that it's workflow is deterministic. SDD (spec-driven) and RDD (receipt-driven) are real state machines behind the scenes: typed phases, persisted state, exact exits. Same input, same lifecycle, whatever model is driving. The model does the work, it does NOT get to improvise the process.

RDD is the one I'm proudest of and it doesn't exist anywhere else: immutable candidate, four adversarial review lenses, one bounded correction max, and a receipt you can verify. Nothing approves itself. When something can't proceed, you get the exact command that fixes it, not a shrug.

And yes, it works with all of them. One sync projects the same contract into Codex, Claude Code, OpenCode, Pi, and more (up to 16 agents). Change model tomorrow, keep your workflow.

Gentle Pi is the same thing baked into the Pi coding agent if you want to see it end to end.

Open source (Totally free and will continue being it), contracts are versioned JSON schemas, there's a CI bench of full journeys. Take a look and tell me what's missing, I mean it: https://t.co/FQFSM2c6pX

[Patrick Vilchez @PvilchezV]

@G_Programming Mucho laburo poca exposicion, atte. no importa cuando lo veas hay una nueva version en horno.

[Gentleman Programming @G_Programming]

Algunas imágenes del nuevo gentle-pi 3.0, alias - Gentle Shell, el nuevo agente propio de gentle-ai basado en pi-agents. Todo lo que ven es hecho nativo:

- Podrás ver el status de tu sub de OpenAI y Claude directo desde la status bar inferior.
- Interfaz totalmente moderna, nativa y super rápida.
- Consumo de recursos MÍNIMO.
- Diff Checker para que puedas ver qué está haciendo tu agente en vivo.
- Manejo NATIVO de sub agentes con posibilidad de abrir su session en cualquier editor de tu preferencia.

Y muchas más cosas que se vienen :)

#opensource #agents #llm #ai #workflow #work #agent #engineering