← Dagens Vibes

Dagens Vibes — 3. juni 2026

Dagens vibe: agent-platformene kravler ud af editoren og ind i kontorlandskabet. Mere Sites, plugins, personlige assistenter — og ja, Chipotle har åbenbart også AI-endpoints. Det er en sektor, ikke en satire.

Fra X-feedet

OpenAI rykkede Codex hårdt mod “arbejdsplatform”: Sites som delbare mini-apps, rolle-plugins og annotations. Det lugter af internal tools for alle — med governance-parfume sprøjtet ovenpå.

OpenAI@OpenAI

Building apps has never been easier. With Sites, Codex can turn your work, ideas, and plans into an interactive website or app your team can explore, use, and share with a URL. Rolling out to Business and Enterprise plans, before expanding more broadly. https://t.co/fF17Y2EzCP

♥ 13.7k↻ 1.2k💬 630🔖 7k
https://x.com/OpenAI/status/2061845949170045346
OpenAI@OpenAI

We’re making Codex more useful for your work by expanding plugins beyond individual tools. These plugins turn Codex into a specialist for a specific role with a single install, no coding required. Codex can access 62 popular apps and 110 skills for work across sales, data analytics, creative production, product design, and public equity investing. openai.com/index/codex-fo…

♥ 3k↻ 267💬 180🔖 1.4k
https://x.com/OpenAI/status/2061887650391625870

Claude Code-sporet handler om bedre agent-orkestrering: baggrunds-forks, API-CLI og selvkontrol. Mindre tryllestav, mere procesmotor.

ClaudeDevs@ClaudeDevs

We've updated /fork in Claude Code /fork now runs a background agent with your exact context (system prompt, tools, history, model) and prompt cache. The result gets returned to your session. /branch (the old /fork) still copies the transcript to a new session you drive. https://t.co/eMyFrk8PfR

♥ 1.7k↻ 107💬 77🔖 797
https://x.com/ClaudeDevs/status/2061947411141169494
ClaudeDevs@ClaudeDevs

We’ve added a CLI for Claude Platform to make every API endpoint runnable from your terminal. Call the Messages API, stand up Claude Managed Agents, pipe results straight into your shell. The ant CLI is well understood by coding agents (Claude Code) using the claude-api skill. https://t.co/t2ruhuAzRH

♥ 4.1k↻ 278💬 152🔖 2.4k
https://x.com/ClaudeDevs/status/2061877343078244459
ClaudeDevs@ClaudeDevs

How do you get Claude Code to check its own work before handing it back? Watch how you can encode your manual checks so Claude closes its own feedback loop: https://t.co/gxastVL4QH

♥ 2.6k↻ 144💬 68🔖 3.5k
https://x.com/ClaudeDevs/status/2061900434722496604

Microsoft gik fuld “enterprise agent”: OpenClaw ind i Windows/M365, Scout som personlig arbejdsassistent og MAI-Thinking-1 som egen reasoning-model. Det er ikke bare Copilot med ny hat.

OpenClaw🦞@openclaw

"You can run OpenClaw inside your company now." Annoucing our work with @Microsoft to bring OpenClaw to the Microsoft and Windows ecosystems. Claws now work securly in the enterprise. https://t.co/3fKhT2mGpC

♥ 3.5k↻ 466💬 242🔖 885
https://x.com/openclaw/status/2061869633624580452
Tom Warren@tomwarren

Microsoft Scout is a new AI personal assistant built on OpenClaw. Scout is Microsoft’s "first real personal assistant," and you can download the desktop app today. Full details 👇theverge.com/news/939713/mi…

♥ 354↻ 37💬 12🔖 188
https://x.com/tomwarren/status/2061870677280624800
elie@eliebakouch

WOW microsoft new "MAI Thinking 1" model comes with a 109 page tech report that looks REALLY detailed, this is amazing https://t.co/VWpxB7VdOb

♥ 673↻ 87💬 20🔖 470
https://x.com/eliebakouch/status/2061877335960281459

DeepMind skubbede AI som forskningspartner med Co-Scientist. Multi-agent hypotesemaskine — præcis den slags ting der lyder fluffy, indtil det pludselig bliver laboratorie-infrastruktur.

Google DeepMind@GoogleDeepMind

We believe AI can be a dedicated research partner to help discover the next breakthrough. Enter Co-Scientist: our latest Gemini-based multi-agent system that can generate, debate and evolve novel hypotheses for complex scientific problems 🧵

♥ 1k↻ 208💬 63🔖 341
https://x.com/GoogleDeepMind/status/2061857539977842793

Thorsten Ball leverede dagens bedste byggementalitet: softwareudvikling er læring, så feedback-loopet er produktet. Det passer ubehageligt godt til hele agent-bølgen.

Thorsten Ball@thorstenball

An internal Slack post I shared with the Amp team a few weeks ago: Building software is learning. registerspill.thorstenball.com/p/building-sof… https://t.co/vrKsLJoQxc

♥ 188↻ 24💬 9🔖 184
https://x.com/thorstenball/status/2061834267240583185

Og dagens cursed sidefund: opencode via Chipotles usikrede AI-endpoints. Burrito-drevet model routing. Silicon Valley bør ikke have lov at være så dumt og så sjovt samtidig.

dax@thdxr

someone made a fork of opencode that routes through the unsecured ai endpoints from chipotle https://t.co/pEFX2Z0E3G

♥ 9.1k↻ 469💬 159🔖 1.4k
https://x.com/thdxr/status/2061828564773740999

Nyhedsbonus

OpenAI’s officielle gennemgang sætter tal på skiftet: Codex har 5 mio. ugentlige brugere, og ikke-udviklere er nu ca. 20% og vokser hurtigere end udviklere. Det er den store historie bag dagens feed.

Codex for every role, tool, and workflowopenai.comhttps://openai.com/index/codex-for-every-role-tool-workflow/

Microsofts Build-overblik samler Scout, Work IQ, MAI-modeller, RTX Spark og agent-sandboxing. Kort sagt: “vi har hele stacken, og den skal styre din arbejdsdag”. Diskret som en blå elefant.

Microsoft Build 2026: Be yourself at workblogs.microsoft.comhttps://blogs.microsoft.com/blog/2026/06/02/microsoft-build-2026-be-yourself-at-work/