Why does this reference Lovable at all? The tagline for this project is "Clone and recreate any website as a modern React app in seconds" and requires a FIRECRAWL_API_KEY in order to perform the necessary scraping.
Lovable is more of a self-contained LLM chat which creates a new react site with optional supabase integration.
I still sometimes have to refer to OpenHands as "formerly known as OpenDevin". It may be limiting in the long run, but it definitely had some short-term punchiness.
It's not really an "open lovable". I mean, there is of course a NextJS repo as part of it, but you need to bring your own API keys to an AI provider, as well as e2b.dev (sandbox, required) as well as firecrawl.dev (web scraping, oddly required).
It's the web scraping more than anything that makes me the least enthusiastic. I get that's the core of their business, but packaged this way, it's much less a loveable-style "build your vision" and more a "quickly copy another site". Just serving as a copy machine is the least interesting use of all this insane compute.
It feels like a common thing in the AI space (moreso than other OSS spaces). People don’t get how trademarks work and that you can’t just name your project after something that’s already popular. I think some people think it’s a clever growth hack.
nope, fire your lawyer. you get a common law trademark by trading under a name. no need to register it. TM registration is just to streamline big corps chasing after lots of small abusers in a wackamole way, that’s all. if the abuse in question is as egregious as “I used their product name, barely modified, to market my own thing, which is not just in the same general market but a literal clone of their product” then a judge is not going to say “Ah but I see they didn’t send in a form to register their trademark, so yeah carry on with blatantly stealing their work lol”
I feel like "also trademark registration is not remotely relevant here" is a distracting statement then. Not necessarily wrong but easy for people to bikeshed over.
Can we run an agent in Github CI? I create a Github issue with my requirements, the CI runner runs Aider scripting mode (it connects to say, OpenAI api) which looks at the comments, maintains a branch, runs tests, takes browser screenshots and then gives its replies.
Why would I do this? It would be better to have an app that turns a react app into a native web app. Speculation rules and page transitions make it possible already.
Reminds of me of tools like https://huu.la, the main difference is newer generations focuses on utilizing LLM to map design to code, versus the older generations focus on using WYSWYG tooling to edit the page directly.
Lovable is more of a self-contained LLM chat which creates a new react site with optional supabase integration.
reply