Ilpo Leppänen

We're starting "What We Shipped," a monthly stream where we'll share our latest tips and talk about new Claude Code releases. The first one is April 7th, come hang out with @dmwlff and me!

Introducing the new dev-browser cli. The fastest way for an agent to use a browser is to let it write code. Just `npm i -g dev-browser` and tell your agent to "use dev-browser"

@bcherny only 👏 for 👏 Team subs 👏 no way

no 👏 more 👏 permission prompts 👏

This is the simple and sole reason why I've been defaulting to voice-use for so long, facing heavy friction for anything NL related using keyboard nowadays

“How do you _” Just talk to it “I really need to _” Just talk to it “My computer doesn’t support _” Just talk to it “I don’t have experience with _” Just talk to it “What should I _” Just talk to it “Does _ make sense?” Just talk to to steipete.me/posts/just-tal…

Background Self-Improvement Loop After your response is delivered, a separate review agent spawns and it decides what's worth remembering and what should become reusable skills, then exits. Hermes gets smarter over time without getting distracted from your work.

i deleted half my Claude setup last week and every output got BETTER sounds backwards, but anthropic's own team just explained exactly why it works. here's the one prompt that tells you what to cut (and you don't even have to paste anything): this is what happens to everyone... you get a bad output, so you add a rule to your skills. "be more concise." next week, another bad output. another rule. "use a casual tone." but a month later, something else breaks. "always explain technical terms." you keep stacking, and it feels productive because you're fixing problems as they come up. but 3 months in, you've got 30 rules piled on top of each other. some of them contradict each other ("be concise" and "always explain your reasoning" are fighting). some of them fix problems that the model doesn't even have anymore. and the model is trying to follow all of them at once, which means it's doing none of them well. it's like handing a chef a 47-step recipe when they only need 12. the extra 35 steps slow the chef down, make them second-guess the parts they already know, and the dish comes out worse than if you'd just let them cook. that's what over-prompting does. anthropic just published a piece on how they build claude code (the ai coding agent). their own engineering team found that their scaffolding was making the ai worse which means your custom instructions are almost certainly doing the same thing. so here's the actionable move... instead of manually reading through your setup line by line, just tell claude to audit itself. if you're in claude's desktop app, claude already has access to your: claude[.]md (the file where your preferences and rules live), your skills folder (where your reusable instruction files are stored), your context files, everything. just open claude code/cowork and say this: — "read my entire setup before responding. check my claude .md, every skill in my skills folder, every file in my context folder, and any other instruction files you can find. then go through every rule, instruction, and preference you found. for each one, tell me: 1. is this something you already do by default without being told? 2. does this contradict or conflict with another rule somewhere else in my setup? 3. does this repeat something that's already covered by a different rule or file? 4. does this read like it was added to fix one specific bad output rather than improve outputs overall? 5. is this so vague that you'd interpret it differently every time? (ex: 'be more natural' or 'use a good tone') then give me a list of everything you'd cut with a one-line reason for each, a list of any conflicts you found between files, and a cleaned up version of my claude.md with the dead weight removed." — one message. claude goes and reads your entire setup, audits it, and comes back with exactly what to cut and why. you don't dig through files, you don't read every rule yourself. it does the whole thing. once you get the results, don't just blindly delete everything it flags. here's the process: 1. read what it flagged and why 2. delete the flagged rules 3. run your 3 most common tasks with the trimmed setup 4. did the output stay the same or get better? the deleted rules were dead weight 5. did something specific break? add back just that one rule the goal is to find the minimum viable setup that gets you the output you want. your ai setup should be getting simpler over time. addition by subtraction baby

GPT-5.4 Pro continues to be the only model of its class. For anything really hard & complex, I throw it into the maw with every bit of context I can think of. More often than not, something very useful comes out. I can't get the same results from Codex or Code or anything else.

Goodbye --dangerously-skip-permissions, hello auto mode

@dsp_ 🎉🎉

No more context bloat from unused MCP servers in your context. Claude, Cowork and Claude Code now load MCP tools on demand. Before and after:

Uni-1 is here! A new kind of model that thinks and generates pixels simultaneously. Less artificial. More intelligent.

Ghostty just surpassed Terraform in stars (my previous most-starred project I started). It took Terraform 12 years to reach 48K. Ghostty did it in 1 year. It's bigger than Terraform in active usage, too. I take it personally when people doubt I can outdo my past. I can take credit for starting both, but not for ongoing development (for the successes and failures). Neither project is a solo endeavor. I'm still extremely actively involved with Ghostty, but there's also a team of a dozen maintainers. Terraform I stepped back and stopped working on it directly like 6 or more years ago. I consider stars a vanity metric and I don't care about it at all except in this narrow case. I'm a super competitive person (in general), but particularly/especially against my past self. There's no one I like "winning" more against than my past. So, this is my one exception for caring about stars.

Introducing Generative TUI Ask anything - get polished dashboards with real data, rendered live in your terminal. 27 components. Streaming. json-render + Ink. npx skills add vercel-labs/json-render --skill ink

Tons of people followed me in the last 3 days, here's my most important video for learning how to work like me. youtu.be/VgR66ybAtdg

we're testing a new version of /init based on your feedback- it should interview you and help setup skills, hooks, etc. you can enable it with this env_var flag: CLAUDE_CODE_NEW_INIT=1 claude would love your feedback!

I put a lot of heart into my technical writing, I hope it's useful to you all. 📌 Here's a pinned thread of everything I've written. (much of this will be posted on the Claude blog soon as well)

@rezoshm @minchoi For those who are seeking youtu.be/gwW8GKwHB3I?is…

@minchoi His recent talk at the All-In podcast was also amazing. Good explanation on how to look at token spend. x.com/rezoshm/status…

Instead of watching 2 hour Netflix movie tonight, watch this OpenClaw FULL COURSE. It'll take you from "installed" to "actually working."

@asidorenko_ I think so too! Running gpt-5.4 in @opencode over long-running threads is satisfying especially when getting functional results but the lack of detail it surfaces by default can easily make you blind to what's happening (or is not happening) in the back..

Codex app/T3 Code/Conductor should copy the file panel from Cursor Glass New agentic IDEs still need a file view, even if a simplified one. Opening another app for this is too much friction. Cursor nailed the ux imo

🚀 Codex 0.116.0 is out! 🔐 TUI: ChatGPT device-code sign-in support 🔌 Plugin auto-install prompts and remote sync 🪝 New hook to block or modify prompts 🎙 Realtime starts with recent thread context Changelog: github.com/openai/codex/r…