Since webcontainer.io is commercial product, i have experiment using loveable to generate code and with backend gptengineer and when running it bolt.diy, it is working. I wonder, why not use gptengineer in backend itself GitHub - AntonOsika/gpt-engineer: Platform to experiment with the AI Software Engineer. Terminal based. NOTE: Very different from https://gptengineer.app. currently webcontainer-io always require assets from webcontainer-io
is there away to host webcontainer-io locally? it seems, it always need to get it from webcontainer-io site?