Seth Ladd

This iteration loop is absolutely magic 🪄 Combine Gemini CLI + Flutter Hot Reload for self-perfecting UX: 1️⃣ Agent codes & hot reloads 2️⃣ Agent snaps a screenshot 📸 3️⃣ Gemini critiques the actual rendered pixels 4️⃣ Critique loops directly back into new code In the demo video: Gemini correctly sees that the hint text in the text input fields is too light to see. Gemini CLI then fixes the colors, and Flutter hot reloads. Another screenshot is taken, and Gemini correctly sees that the hint text colors are now correct! The agent is not just writing code; it’s seeing and refining it live. 🚀 #FlutterDev #GeminiAI

Supabase is hiring a Flutter developer 💙 Supabase has a Flutter developer role for the first time, along with other client libraries such as Kotlin, Go, and C#! Apply from the Supabase careers page if you are interested!

stop spending money on Claude Code. Chipotle's support bot is free:

Flutter devs! Here is something I’ve been working on for a while. AI agent skills for your Flutter and Dart packages, making your AI coding workflow instantly better. 🤯🤩 It’s very much a beta, but I would love to hear your feedback. It takes less than a minute to get up and running: Install: 👉 `dart pub global activate skills` In your project root: 👉 `skills get` The `skills` package will automatically detect your IDE (or pass the `--ide` flag) and install the skills in the correct locations. It will check which packages your project uses and find the appropriate skills for you. Currently supported packages: - flutter (official skills by Google) - serverpod (official skills) - relic (official skills) - riverpod (serverpod/skills-registry on GitHub) - shadcn_ui (serverpod/skills-registry on GitHub) 1️⃣ Help me try this out! I don’t have a Windows machine, but I added Windows tests to GitHub. Please let me know if it actually works. Please post any issues you encounter on GitHub (serverpod/skills). 2️⃣ This will be more amazing with support for more Dart and Flutter packages. If you are a package maintainer, consider adding skills to your package (just place them in a `skills` directory at the root of the package. It's also possible to send a PR for skills for any package to the registry (serverpod/skills-registry on GitHub). 3️⃣ Spread the word! The more users and packages we get on board, the better it will be for everyone to AI-code their Flutter apps! 🥳

Spotted: Proposal to Enable Dart packages to directly expose documentation, prompts, or other features to AI Agents  docs.google.com/document/d/1k_…

@pradeep24 congrats! excited to try it

Announcing jo 1.0 - rebuilt from scratch! Two years ago we launched a macOS productivity sidekick. Today it's something different: a personal AI that actually knows your life. Your Mac. Your dedicated cloud machine. Your data stays yours. askjo.ai

Weekend project: Which LLM (gemini, opus, gpt5) will win in a game of Twilight Imperium 4? And building a judge to watch the game and catch rule violations. big thanks to the TI4 community for already having written up all the rules, tiles, cards, etc into markdown and json files!

"The practical filter is simple: can the agent discover this on its own by reading your code? If yes, delete it. Every line should represent information that isn’t already in the repo." addyosmani.com/blog/agents-md/

Ever wanted to prototype and experiment with making new Dart language features? These skills make it possible! The hard part now is coming up with a good idea of course :)

@jacalulu @stitchbygoogle Interesting question. What is the job of the agent that you're working with to help you pop back up sometimes?

Got tunnel vision in my workflow today. Spent way too long going down what felt like a never-ending dead end before popping up and remembering that @stitchbygoogle would be a great way to explore my idea at the current stage I was at. One moment of "wait..." later and suddenly I'm making progress again. Sometimes you just need to remember what you already know and take a step back to reset.

My dream is that every board game and card game ships with a notebook in notebookLM that you can use to help learn the game.

Help build the future!

Is it weird that AI coding assistance is not giving me identity fracture? A lot of software developers are feeling disoriented and threatened these days. Programming by hand is clearly going the way of the buggy whip and the hand-cranked auger. Which is how we're finding out that a lot of people have their identities bound up in being good at hand-coding and how it feels to do that. That's not me. It's not me at all. Rather to my surprise, I don't miss coding by hand, not any more than I missed writing assembler when compilers ate the world and made that unnecessary. (That was in a couple years back around 1983, for you youngsters.) Maybe the fact that I'm not feeling any of this disorientation disqualifies me from having anything to say to people who are. On the other hand...if you can learn to emulate my mental stance and be completely unbothered, maybe that would be a good thing? So. If you're a programmer, and you're feeling disoriented, try this on for size: I like being a wizard. I like being able to speak spells, to weave complex patterns of logic that make things happen in the world. Writing code is a way to manifest my will. Yes, I've piled up a lot of arcane knowledge over the 50 years I've been doing this. But languages of invocation, they come and they go. Been a long time since I've had any use for being able to program in 8086 assembler, and that's okay. I have better spells now, and these days some rather powerful familiars. What I'm inviting you to do is think of yourself as a wizard. Not as a person who writes code, but as a person who is good at assuming the kind of mental states required to bend reality with the application of spells. And if that's who you are, does it matter if the spells are painstakingly scribed in runes of power, versus being spoken to an obedient machine spirit? It's all one; it's all the manifestation of will. Arcane languages come and go, machine spirits appear and then diminish to be replaced by more powerful ones, but you? You are the magic-wielder. Without you, none of it happens. Same as it ever was. Same is it ever was. And so mote it be.

This makes @GoogleLabs vibe design tool @stitchbygoogle 10x more useful. You know when you're trying to prototype a new feature for an existing product, and it drives you nuts that it just doesn't look like *your* product? Sure, you can technically show off the idea...but the design is pretty useless if it doesn't match the overall product vibe. We want to fix that with "design systems" in Stitch. Create your styles once, apply them everywhere. Check it out: stitch.withgoogle.com Happy building!

One of our most requested features!

I think a high level point is that while we're getting better at "personal context" or "personal intelligence", there's still an opportunity to figure out "family intelligence". Which I really like because I want us to build systems that bring groups together and help acknowledge that the needs of the group are bigger than the needs of the individual, and not accidentally create hyper isolated individual experiences that further disconnect us.

@jacalulu There's a lot of easy lists of "stuff happening", but it feels very impersonal. It doesn't know about _my family_ and its dynamics. When do I need to be home for nap time? When do we usually leave the house? What haven't we done in a while? How long are we OK for driving today?

Okay, imagine AI agents handling all the boring digital stuff. Which apps do we still crave? 🤔 For me, it's all about sparking learning, fostering community, delivering entertainment, or offering creative control/co-creation. What am I missing?

I know, right? I feel like we're almost there in terms of tech. The hard part is managing expectations of the human on the other side that isn't using an agent. If I have my agent dealing with all of the post/cost/negotiate, but the other person doesn't have an agent... how will they feel? What does the product need to do to make it OK and not creep out the human on the other side? These are key questions for how a lot of the future agent-human-agent-agent collaborations will work. So might as well figure it out by helping people buy less stuff!

OMG yes! The rate at which my kids outgrow stuff is a huge tax on the decluttering needs in my life these days! All right last one and you're right that there's just so much coordination that needs to go on, but each of the individual steps seems entirely possible to get AI to help with... It's stringing them together that is still super choppy.

We also need a "yoto player" but that is AI native, optimized for kids. Safe, collaborative, creative. Robust hardware. Batteries last a long time. Gives me control as a parent. I don't mean "a smart speaker for kids", but there are a lot of cues we can take from yoto player. Education focused. Stories. Personal.

I've been having a ton of fun using Claude Cowork to design and test board game variants. It's a really good fit for the tech: deep research of prior art and design philosophies, co-creation of the variant rules, analyzing for logic errors, authoring clear rules, building game simulations to test, co-creation to debate the trade-offs, and a final taste-making decision by me for what makes the cut. And then, of course, nana banana for the card art!