Sebastian Huckleberry

I'm now using web.runme.dev as a frontend for Agents built using OpenAI's agents API. For long running agents (e.g. a chaos agent) the UX is doc per agent session with widgets rendering the conversation and interacting with the agent.

Here's how I'm using Codex & @runme.dev to turn the crank at OpenAI. learn.chatgpt.com/blog/automat...

Ha. #LLMs are bad jokers and a tough audience at the same time. This one scored 78% 🤷‍♂️.

A lint task once wandered too wide, And dragged the whole codebase inside. Now two files compile, In isolated style, With Docker complexity denied.

The Build your Own Agent industrial complex has you believe only home-made AI agents need evaluation. Nope. Behavior you deploy into AI agent harnesses need testing, too. This linkedin post just struck a major nerve. www.linkedin.com/posts/sebast...

If my “World Cup picks report” #AI agent skill had a theme song, it’d be this one. Eval results: world-cup-picks-report-evals.sourishkrout.workers.dev Skill and evals: github.com/sourishkrout...

FIFA please license a broader music catalog. I’m done Living On A Prayer.

“F*** your skills” is the joke. “And mine too” is the point. We stopped shipping software without tests. But we’re still putting production-grade behavior into agents on lucky runs, screenshots, and folklore. Use `runme eval` with tests and results in the repo: runme.dev/blog/runme-e...

I still hate you, #FIFA, but I did love everything about this ⚽️ soccer game yesterday. #CURGER

Soft #agentskills.io are just prompts. Prove me wrong.

Your diff shows what changed. Your trajectory shows why it worked. #CodingAgents Keep both: visr.dev/changelog/ke...

don't remember the last time I ran `git grep ...`. it's all about asking the agent these days.

The cost of product launches has collapsed drastically. Yesterday, visr.dev had: - no blog, - no content, - no design, - no video clips, And, we had yet to polish the actual product feature. It dropped today and it ain't AI-slop: visr.dev/changelog Used to take ~2wks for a small team.

Shipping Visr v0.1.0 today. It’s early, it’s beta, and it’s aimed at one thing: Keeping the 🤖 Clankers on Rails by closing the 🔁 context loop inside AI coding harnesses. Feedback welcome: visr.dev/changelog/bo...

Tonight we congratulate our friend Stephen Colbert and his writers, staff and crew for eleven years of excellence. Watch Colbert’s finale tonight, and then never watch CBS again.

Didn't think my repo was so esoteric that @posthog.com's AI integration wizard gets confused. Boy, was I wrong.

Trude Kernig, back in business. My OpenClaw forced my hand. I’m fully migrated to Codex. Claude didn’t lose on taste. Anthropic degraded out of first-choice status. At parallel coding speed, throughput wins.

Gearing up for the early Visr.dev release. Simple goal: help AI agents stop making the same mistake twice. We’re turning past runs into reusable skills, checks, and evals for agent harnesses. If your team's building with agents, this is probably a problem you’ve already felt. Waitlist is open.

Remember how GitHub's subline was "Social Coding"? Where's the new thing... "Agentic Coding"?

#OpenClaw is a fascinating case study of modern software: coding can scale in parallel, but reducing complexity takes time and a big-picture view. That is why I do not buy the idea that agents need all context upfront. Development is a fast feedback loop, not a one-shot prompt.

It's wild to see the weirdest rate-limits hit you once you dial up agentic coding concurrency.

Memory getting in the way?

TIL that "minime" means "no" in Latin.

One from horse-less carriage category. Against @jeremy.lewi.us's efforts Golang must no be particular popular at OpenAI 🥸.

Jumped on the #openclaw hype train Nailed choosing its name

*just* learned that the slang word for AI agents is clankers. i feel old.

if agents are our OS/shell, prompts our code, and MCP/tools are APIs, would we refer to agentskills.io as binaries? if they are, why aren't we treating them that way?

Goofy and Ted Lasso are the same person. Change my mind.

It's always fun to see a big traffic spike hit @runme.dev's website. Today @console.dev featured Runme in their newsletter. Thanks for the shout out and also appreciate the likes / dont likes.

this liquid glass is going to take some serious "getting used to"

Hot take: Terminal agents are a dead-end. While terminals, the TTY interface, is key for operators, it's backwards to lock the UX into the terminal. www.youtube.com/watch?v=MXOP...

How is it possible that Azure takes longer than AWS to bring up a Windows VM? 🙈

We have 🤖 AI notetakers in meetings but continue to silo know-how every time we close terminals. Not just the how but also the why and what. Sign up for beta access to visr.sh: it's like as if Granola AI and #tmux had a kid.

What if it isn't productivity gains in every day things that's driving ChatGPT? Perhaps ChatGPT's popularity is a natural extension of the para-social relationship folks have developed with tech. Facebook's site helps you "connect" with your friends. Now the "site" has become your friend?

Deploy a vanilla react app (e.g. with vite) with an /api folder containing some cloud functions to @vercel.com. Forgo the @nextjs.org dance. Thank me later.

Mezgroni 🥃? There’s got to be a better name for a Mezcal Negroni. One week sample from the barrel 🪵 is excellent 🤯.

You and I are gonna live forever 🎶 #oasis

Sometimes Javascript is just killing me.

Name one good reason why I shouldn’t keep my iPhone in Low Power Mode permanently.

I have to give I to the backpacks made by peakdesign.com. They don’t just look sleek. They continue to save the day. Look how effortlessly I strapped it to the motobike on my impromptu Provonce 🇫🇷 road trip. Viola!

Marseille is an excellent place to visit. It’s vibrant, eclectic, beautiful, and full of interesting cuisines 🍲🐟. Can’t say no to a Bouillabaisse even on a hot August day!

Monaco 🇲🇨 is a reverse Las Vegas. In Vegas the glitz is fake and the fun is real. Draw your own conclusions 😆.

The best the Liguria’s Genova has to offer. - Fish with ground garbanzo bean flour chips - Pesto Genovese, invented here Yummys 😋

AI-powered coding at 34,000ft in #Cursor: no problem ✈️. Contrary to the laptop charger that keeps falling out of outlet 🤦🏻‍♂️.

If the rumors are true and Coca Cola will release their own cane sugar version to compete with Mexicoke will we call it Coke Cane? Asking for a friend.

Released software from the backseat of a driverless Waymo ✅