That’s really cool. It’s one of those ideas that seems obvious in retrospect once you see it.
What’s your moat? Especially against companies like Webflow or Vercel who could implement your “product as a feature.” Have you thought about partnering with those sort of companies to white label your solution embedded into their apps?
Thanks, really appreciate it. Our moat is that we're completely open. Matter of fact, even the interface to connect other agents is open source. We want to keep being compatible with all kinds of frameworks and tech stacks.
oh and as stupid as it might sound, but: we're confident that we can simply be better. Our roadmap is full of things that go way beyond the current capabilities, but everything is focussed on one goal: Make building production-grade (and pretty) frontends as easy as possible.
Sounds cheesy, but the focus on really really nailing this one thing sets us apart we think.
This is what I've been looking for! Tried self hosting the bolt.new editor to achieve something like this and it was terrible. Also interested in hearing about the SaaS you were building, I recently built something in that niche coincidentally drafted at first by v0.
Ha, this project has had me thinking about that sector more, I'll try and make an effort and go for it in the future, and congrats you have built something truly useful here.
I was building something for a single logistics company; they'd handle clearing and forwarding for their clients as well as transportation if it was required, covered everything from invoicing to fleet management, outsourcing jobs, It's evolved quite a bit considering the initial premise was simply creating quotes and invoicing, maybe It'll end up being a bigger deal in the future
Sounds like the experiences we've had in the industry - even though they're all 'logistics companies', their workflows and requirements differ greatly, even within a niche of a niche. This gave us a hard time building a scalable software that would be useful for as many companies in the industry as possible
It works with everything that runs in a browser - as long as the dev app runs on localhost and the codebase is local as well, we can integrate with it!
Ugh, looking for something like this for Front-end Mobile Dev (React-Native, Swift etc.). The closest thing I've found is https://www.magicpatterns.com, but that doesn't work on existing repos. Keep that in mind!
Thanks! There are also some other tools out there that give 'vision' to your coding agent via MCP, but we figured that prompting the agent directly by clicking on elements is the most intuitive way to interact with it.
Yeah the video is amazing. I like it when some magic feeling is added back to AI. We got so used to chatGPT the dopamine hits are rarer. But chatting to your app to change the code rather than merely use the app is magical. For now... (need a new sparkle tomorrow lol!)
Haha, u're right, I recently had a chat w a friend about how high our threshold for excitement has become. A few years ago, a new iPhone release would have meant the world
It's so funny how many people tell us that - it's one of the obvious ideas someone just needed to start with. What kept you from building the first version?
Are you using one-time mail providers? Otherwise, please check the spam folder of course. If nothing helps, feel free to join Discord (https://discord.gg/DjMqMnUVCh) and we'll debug this together.
v0 and lovable have way too much hype, I think what you're doing with stagewise (i.e. being dev-first instead of optimizing for non-developer users) is the real long term use of AI.
That’s really cool. It’s one of those ideas that seems obvious in retrospect once you see it.
What’s your moat? Especially against companies like Webflow or Vercel who could implement your “product as a feature.” Have you thought about partnering with those sort of companies to white label your solution embedded into their apps?
Thanks, really appreciate it. Our moat is that we're completely open. Matter of fact, even the interface to connect other agents is open source. We want to keep being compatible with all kinds of frameworks and tech stacks.
oh and as stupid as it might sound, but: we're confident that we can simply be better. Our roadmap is full of things that go way beyond the current capabilities, but everything is focussed on one goal: Make building production-grade (and pretty) frontends as easy as possible.
Sounds cheesy, but the focus on really really nailing this one thing sets us apart we think.
This is what I've been looking for! Tried self hosting the bolt.new editor to achieve something like this and it was terrible. Also interested in hearing about the SaaS you were building, I recently built something in that niche coincidentally drafted at first by v0.
Haha, cool to hear!
We were tinkering with a transport management system because the industry was full of legacy software.
But before we were able to battle-test something, stagewise took off.
Still think that it's an industry worth going for (at least in Germany)
Ha, this project has had me thinking about that sector more, I'll try and make an effort and go for it in the future, and congrats you have built something truly useful here.
What were you building? And did it evolve/ has been used by someone?
I was building something for a single logistics company; they'd handle clearing and forwarding for their clients as well as transportation if it was required, covered everything from invoicing to fleet management, outsourcing jobs, It's evolved quite a bit considering the initial premise was simply creating quotes and invoicing, maybe It'll end up being a bigger deal in the future
Sounds like the experiences we've had in the industry - even though they're all 'logistics companies', their workflows and requirements differ greatly, even within a niche of a niche. This gave us a hard time building a scalable software that would be useful for as many companies in the industry as possible
Would this work with Ruby on Rails apps that use erb templates? Or does it have to work with JavaScript frameworks on the front end?
It works with everything that runs in a browser - as long as the dev app runs on localhost and the codebase is local as well, we can integrate with it!
This is pretty wild. Very clever idea, kudos/congrats.
Thanks! Would love to hear product feedback since there's still a ton we need to improve
VanillaJS support pls then can be used in all sorts of webapps etc not using frameworks
Been using this for a few weeks now, and I love it. I'd love it if yall could add some more design controls!
Have you tried the stagewise agent so far? If so, what are the core problems the agent still has?
Nice to hear that! What exactly do you imagine when talking about design controls?
Ugh, looking for something like this for Front-end Mobile Dev (React-Native, Swift etc.). The closest thing I've found is https://www.magicpatterns.com, but that doesn't work on existing repos. Keep that in mind!
The whole 'vibe-coding' space seems a bit underserved for mobile or am I wrong?
Cursor/Claude works decently enough for react-native.
Would love some models optimized for Swift and also closing the loop with the simulator would be clutch.
Will do so!
Does it work with Node/JS/TS front ends only, or can it work with other languages and web frameworks?
It will work as long as the project is a web project and runs in the browser. The agent is a pure js snippet and can be injected into any web app
Smart solution. My todo list had getting Claude Code somehow to "look at the ui" but looks like you have prefabbed this sort of thing for us.
Thanks! There are also some other tools out there that give 'vision' to your coding agent via MCP, but we figured that prompting the agent directly by clicking on elements is the most intuitive way to interact with it.
Yeah the video is amazing. I like it when some magic feeling is added back to AI. We got so used to chatGPT the dopamine hits are rarer. But chatting to your app to change the code rather than merely use the app is magical. For now... (need a new sparkle tomorrow lol!)
Haha, u're right, I recently had a chat w a friend about how high our threshold for excitement has become. A few years ago, a new iPhone release would have meant the world
Cool, I had a very similar idea recently, awesome to see it live! Congrats!
It's so funny how many people tell us that - it's one of the obvious ideas someone just needed to start with. What kept you from building the first version?
How do I make an account? There is only an option to Sign In. I entered my email, it said that you sent an email, but there's nothing.
Are you using one-time mail providers? Otherwise, please check the spam folder of course. If nothing helps, feel free to join Discord (https://discord.gg/DjMqMnUVCh) and we'll debug this together.
Hey, I received it, it just took like 10 minutes, I use fastmail
awesome. hope you enjoy
Did the agent work well after the mail came through?
I think it ran out of credits before it could finish one task.
Available plugins: @stagewise-plugins/angular
Agent server initialized successfully
Stagewise is running on http://localhost:4200
Proxying app from port 3100
Opening browser...
Credits: [ ------------------------------ ] 1.99/2€
Credits: [ ------------------------------ ] 1.97/2€
Credits: [ ------------------------------ ] 1.95/2€
Credits: [ ------------------------------ ] 1.92/2€
Credits: [ ------------------------------ ] 1.89/2€
[multiEditTool] Editing file public/i18n/en.json (30679 bytes)
[multiEditTool] Edit 0: Replaced 1 occurrence
[multiEditTool] Edit 1: Replaced 1 occurrence
[multiEditTool] Edit 2: Replaced 1 occurrence
[multiEditTool] Edit 3: Replaced 1 occurrence
Credits: [ ------------------------------ ] 1.86/2€
Hmm, the initial credits are usually enough to do 4-5 major tasks and a few more smaller ones. Was the log with 1.86/2 credits the latest?
Oh maybe it's not that. It showed me "task failed" in the UI, and I saw 1.99/2 in the console, so that's what I assumed.
I'm gonna give it another try a bit later.
Oh okay, seems like there was another issue then, I'll debug. Happy to see a run without issues later
Good luck!
The Sign In option also serves as Sign Up.. I'm having a look at the email issue now!
Cool!
How does this work for non-UI code?
Do you mean for UI-related changes that also include modifying the backend, or do you mean pure backend changes that don't involve UI at all?
v0 and lovable have way too much hype, I think what you're doing with stagewise (i.e. being dev-first instead of optimizing for non-developer users) is the real long term use of AI.
thanks for the kind words! If you have ideas or wishes for next steps, feel free to share.
existing codebases... so long as they are web applications?
Yep, that's correct. stagewise only works with web applications.