Dagens Vibes — 25. august 2026

Introducing Headlong, an open source microharness for persistent agents: self-guided agents that think continuously. Most agent harnesses are reactive: you send a task, the agent completes it, and then it sits frozen until the next request. Cron jobs and heartbeats wake it up to run a checklist and put it back to sleep. A Headlong agent is never asleep. It keeps generating thoughts about whatever it decides is interesting, in a self-guided loop inspired by human inner monologue. Your message doesn't start a session. It's one more observation that lands in the agent's thought stream, and the agent decides if and when to reply. Headlong is built on the idea of persistent agency: continuous inner thought generation between external interactions. The agent sets its own interests and priorities, comes up with its own projects, and sometimes pings you unprompted with progress. To keep our prototype as simple and small as possible, we implemented Headlong as a microharness: a complete agent harness in under 10K lines of Bash, organized as a handful of small executables. It includes a loop that generates the next thought, shellm (a recursive language model written in Bash), a trajectory stored as a DAG of jsonl files, and context as a projection of that trajectory. We've been running one Headlong agent internally at Laude for several weeks. The whole team talks to it over Slack and Telegram, and every conversation lands in its single stream of thought. It works in its own fork of Headlong and we've pulled over 50 of its commits into main. One night, with nobody talking to it, it went back to check whether a recall process it had built was actually wired into its mind, found that it wasn't, diagnosed and fixed the bug, and verified the fix end to end. 48 minutes, no human asked for the fix or was in the loop at any point. Every step is a timestamped line in its log. Things broke too, and we wrote those up. Background thinking costs us $1 to $2 an hour, our agent stopped its own service three times by accident, and self-delegation died on day one. Details in the post. One line installs everything and starts an agent. Use a dedicated sandbox and spend-capped API key; it runs real shell commands and thinks around the clock. Headlong is research software, be careful! curl -fsSL https://t.co/P2Q5Ch8C4e | bash Launch post: https://t.co/HhWULUTkpH Repo: https://t.co/U4JunpSTYt Headlong is a @LaudeInstitute / MIT collaboration.

https://x.com/andykonwinski/status/2091990178638496195

last time @ryancarson was on the pod, over a quarter million people watched the episode he's back - and we're talking cloud agents, @DevinAI, and our dirty little secret (we're both hiring humans!) in this very special how i ai, ryan and i cover - what he learned spending $20k/mo on devin - why he still keeps a paper to do list - how he keeps hist 15 active threads organized - his QA loop for 40 PRs/day plus - he gives us a peek into how he thinks about interviewing people now: "just let them cook and do a screen share." a huge ty to our amazing sponsors: 💼 @WorkOS - Make your app enterprise-ready, with SSO, SCIM, RBAC, and more: https://t.co/7TkTWqVCqv 🪙 @Jira AI SDLC—Get your tokens’ worth with Jira: https://t.co/lMeieiu2Xe Full ep on YT: https://t.co/GXeMbmK7Yx

https://x.com/clairevo/status/2091927891353498017

Codex has become a real game engine... If you take the time to make it become one. You can do EVERYTHING in it. And I think I've cracked the perfect custom UX? I'd like to show you how I use it. See video below 4 steps: - Create a custom tool you need, like one to generate new buildings that matches you game art direction - Ask Codex to open this tool in a worktree (important) - Attach the browser view of your tool in the Thread - Use pinned thread to have easy access to the tool And there you go: you have what I call a "codex-augmented" tool, which it's incredibly powerful because - the tool will speed up your development immensely AND also: - you can improve the tool as you use it with codex For example, let's say I want to add a new button to my 3D building gen tool so I can generate a thumbnail of the 3D building. - I click on "Building Lab" - I prompt Codex "add a button to generate thumbnails" - I have my new button - I can use the tool right away in the pinned Thread You can then use the /yeet skill to commit, push and create a PR with the changes you made to your tool. I have tools for everything - To manage my tasks in Linear - To do content / social media / steam marketing - To localize the game in 9 languages - To manage sounds, musics and voice lines - To generate and fine tune buildings - To create and animate procedural zombies - To tweak environnement/decor and map design - To manage my analytics - To simulate late game perf I'll be honest it's incredibly powerful and easy!

Nicolas Zullo@NicolasZu

Here's a follow-up, this time with a more ambitious building (+animation, yes!) and with GPT-5.6 Sol (High), in the same thread. The prompt: "Use the ImageGen skill to generate a low-poly 3D mining rig, then use img2threejs to build it. Allow me to animate it with a button." https://t.co/EWk06Rz9O1

https://x.com/NicolasZu/status/2091765370499600520

Spencer is delivering on our promise for the fully agentic OS of The Future. This is even running off a local Qwen 27B model! We'll be shipping a version of this in Omarchy 4.1. GET READY TO ACCELERATE! 🤖🚀

Spencer Bull@SpencerGBull

Using an "Agentic Ready OS" is changing the way I use computers… Today I'm releasing the first alpha of the OmaPilot plugin for @OmarchyLinux and looking for more testers to iron out the Omarchy harness. https://t.co/60cjV8YRsc

https://x.com/dhh/status/2091758798704677048

I'm officially Linux pilled Over the weekend I installed Omarchy on my old gaming computer Since then I've absolutely fallen in love with using computers again Rough notes on why using it is so great, and why I think you need to give it a try: • SUPER performant. Runs incredibly well on an old Dell Alienware computer from years ago that was just sitting in my closet collecting dust • No floating windows just makes sense. By default, all windows in Omarchy auto tile rather than float. Every time you open an app, it splits the screen further automatically. I find it impossible to go back to floating windows on Mac now. Feels archaic and sluggish to manage • No bloatware. Windows for me has become completely unusable the last few years. Copilot, ads, and all the extras that manufacturers include with it has made it the most unpleasant experience in computers. Omarchy comes with none of that. • Feels built for AI. Basically every harness including Codex and Grok Bot have native apps for Omarchy. • Built for productivity. Switching workspaces and using the keyboard for basically everything just feels snappy and quick • Makes computing fun again. Something about the whole experience is just FUN. Learning all the new key bindings, making my set up personal and my own. I got home every night this weekend just looking forward to getting back to my computer. HIGHLY HIGHLY HIGHLY recommend trying Omarchy out for yourself. You don't need to buy a new computer for this. It can run on really old hardware. Just grab a laptop or desktop from your closet and set it up. Ask your agent how to flash a thumb drive so you can install it. Codex basically did the entire setup and installation for me The learning curve is really small. As long as you have an agent up and ask it questions any time you're confused, it should be able to help you out. And if you install tailscale and connect it to your private network, your agents can even control it from other computers. Altho it's not necessary, I'm going to pick up a Framework laptop to use this. Need to take this on the go. Excellent work @dhh

https://x.com/AlexFinn/status/2091938648610816478

Pi 0.84.3 is out! The main things to point out is that you can enable powershell on windows now and it should start much quicker on most systems due to changes in bundling. https://t.co/u7qCrBof5B

https://x.com/pidotdev/status/2091846321472758093