garyshary

FBM #2496

[Codez @0xCodez]

https://t.co/eJB3TgyJwV

[Hyperagent @hyperagentapp]

Fable 5 on Hyperagent is producing the most creative, ambitious work we've ever seen from our agents.

They're self-improving for hours towards open-ended goals. Visual reasoning has spiked noticeably. Outputs are consistently higher quality than Opus, occasionally at lower cost.

5 of our test cases below vs. Opus 4.8 👇

1. Visualize all asteroids in the solar system from NASA data
2. Design a site plan for a 100 acre fitness retreat
3. Reconstruct Apollo control panels from technical PDFs
4. Simulate the supply chain for World Cup jersey sales based on match outcomes
5. Show the effects of solar flares on aurora

Fable 5 is now available on Hyperagent.

[Rahul @sairahul1]

Anthropic and OpenAI are both telling engineers to write loops.

Not prompts.
Not agents.
Loops.

That is not a coincidence.

When the two most important AI labs on the planet independently converge on the same pattern — that is a signal worth paying attention to.

Most engineers are still thinking in terms of single calls.

Input → model → output.

The engineers winning in 2026 think in cycles.

Output becomes input. The model evaluates its own work. The loop runs until the result is right.

This is the complete breakdown of what loops are, why they matter, and how to build them ↓

[Elon Musk @elonmusk]

Grok Imagine on Vercel

[garyshary @garyshary]

Pathetic

[Xohi⭐️⭐️⭐️🇫🇷 @Xohi1260]

When a picture says a thousand words
Andreeva vs Kostyuk 6-1 6-3
Bravo Andreeva 🇷🇺 👏 👏

[Alfred Newmann 🇧🇷 @pmarcelo1972]

Então deixa eu ver se entendi...

O presidente chama Trump de "imbecil", diz que não tem medo dos EUA, afirma que o Brasil não aceitará o tratamento americano, declara que Trump não é o "imperador do mundo", acusa o governo americano de fomentar conflitos, chama Marco Rubio de "latino-americano frustrado", diz que o Brasil não vai chorar se os EUA quiserem problemas e ainda afirma que ninguém deve ter medo dos Estados Unidos.

Mas, se amanhã surgir alguma sanção ou retaliação contra o Brasil, a culpa será de Flávio Bolsonaro.

[Dami-Defi @DamiDefi]

INSTEAD OF opening Obsidian just to dump more notes tonight.

Spend 30 minutes studying these workflows.

12,900+ GitHub stars in under three months. Free.

The ones who connect plugins, workflows, and Claude will turn their vault into working memory.

[Mauricio Marcon @Maubmarcon]

Imaginem o Bolsonaro sugerindo enforcar o Lula e o Moraes. Ia ter até plantão na Globo.

[Prime Video Sport France @PVSportFR]

🇧🇷 João Fonseca et son idole « Guga »

#RolandGarros

[IMPERATOR @IMPERATORAUS]

This is actually the perfect analogy.

[CyrilXBT @cyrilXBT]

Stop sleeping on AI

Obsidian + Claude Code = your own JARVIS.

Takes 1 hour to build.

Most people will scroll past this and stay unproductive.

The ones who stop and build it will never work the same way again.

[ClaudeDevs @ClaudeDevs]

We’ve shipped a security-guidance plugin for Claude Code that helps identify and fix vulnerabilities as you’re writing code.

Available for all Claude Code users. Install from the plugin marketplace (/plugins).

[darkzodchi @zodchiii]

Anthropic engineer showed how one person can run 5 AI agents, that code, test, review, and deploy at the same time.

In 30 minutes they built the whole thing live in one session.

Here's what they cover:

> when to use one agent vs a full team
> how to split work so agents don't step on each other > the exact framework for deciding what each agent handles

that's exactly why, I put together a guide on building agent teams that actually work.

full guide in the article below 👇

[Elon Musk @elonmusk]

Grok foundation model V9-Medium (1.5T) has finished training. Evals look good. A lot of Cursor data was added in supplementary training and there is more to come.

Fine-tuning is underway and reinforcement learning begins in a few days. 2 to 3 weeks to public release.

This will be a major improvement over the 0.5T v8-small that currently serves all Grok production traffic, especially for difficult coding tasks.

[Indu Tripathi @InduTripat82427]

HERMES AGENT IS TAKING OVER FAST.

And most people still have no idea what’s happening.

If you’re still stuck testing basic AI workflows manually, you’re already behind.

140k+ GitHub stars. Top-ranked on OpenRouter. One-command setup. Persistent memory. Agents that evolve over time.

This beginner walkthrough explains the entire setup in under 1 hour:

[AYi @AYi_AInotes]

Damn，这个必须卧槽一下了，Karpathy 的 CLAUDE.md 只有 65 行，
居然能把 AI 编程准确率从 65% 拉到 94%，

以22 万星标登顶 GitHub 趋势榜，而且绝大多数开发者还没读过，

里面没有一行奇技淫巧，4 条规则全是同一个方向：

→ 先想清楚再写代码
陈述假设，不确定就问，杜绝猜测

→ 从最简方案入手
只写能解决问题的最少代码，不加任何多余抽象

→ 像手术一样精准修改
不碰与需求无关的代码，每行改动都对应明确要求

→ 以目标驱动执行
写第一行代码前，把模糊指令转化为可验证的成功标准

65 行，4 条规则，每一条都在对抗开发者“先写再说”的本能，

本质上是把慢下来这件事变成硬规则，

趁别人还没保存，赶紧先存好 👇

[teKa @teKa088]

Anthropic acaba de resolver el problema que frenaba a las empresas grandes con IA agéntica:

Antes: tu agente necesitaba acceso a tu MCP server → tenías que exponerlo a internet público → área legal decía no

Ahora: MCP tunnel crea un canal seguro vía tunnel .anthropic.com → el agente llega a tu servidor interno sin que salga nada a la red pública

Tu firewall, tus políticas, tu perimeter. El agente opera adentro.

Self-hosted sandboxes (beta pública) + MCP tunnels (research preview) = Claude Managed Agents corre en tu infraestructura

[Charly Wargnier @DataChaz]

this guy literally breaks down the exact Claude + Obsidian playbook to build your own AI employee in a single afternoon 🤯

[Anatoli Kopadze @AnatoliKopadze]

Godfather of AI: "If you sleep well tonight, you may not have understood this lecture."

This 47-minute lecture is the best thing I saw about AI in the last few months.

It will definitely help you understand how it actually works and where it's going.

Geoffrey Hinton built the neural networks behind every AI alive, then quit Google to warn the world about it.

The part nobody wanted to hear:

> AI is already developing abilities its creators didn't intend
> in most cognitive tasks it's already ahead of us
> the question is no longer if it surpasses us but when
> the only decision left is which side of that line you're on

Right now the average person opens Claude, types something, gets an answer, closes the tab.

They think they're using AI. they're using maybe 10% of it.

I went through his entire lecture, built a practical system from what he was describing.

18 steps to actually use Claude the right way, with copy-paste prompts that work today.

Full guide in the post below.