> Photoshop-, After Effects-, and Blender-class video tools
Look, you can’t just write this. When you say something like this, you set the bar extremely high, and open yourself up to (totally fair!) criticism if Photoshop or After Effects gurus point out entire classes of use cases you aren’t handling.
If you want people to talk about what your tool is good at, you need to stop inviting comparisons to titans like this. It will take you a long long time to build something as fully featured as them, and there is plenty of room for agent-focused tools that aren’t at that immense product scale.
Info-dense, completely unreadable README.md file. Only the line "No Puppeteer, no Chromium IPC, no `page.screenshot()`" actually allowed me to figure out what this is about: creating video demos of your website (and probably other software as well) with an AI agent, rather than going through the conventional video-editing process.
It's cool that so many people are building useful tools with LLMs, but for the love of ones and zeroes, people! PLEASE write your README by hand, instead of handing it off to the LLM with no human involvement! You know how unreadable LLM prose is! Why do you make that the first thing people see when they see your repo?
I mean, I'd even be happy(-ish) with a README that starts out "Apologies for the LLM text below, but this is still in heavy development, and I haven't gotten around to writing a proper README yet. In the meantime, though, here's something the LLM slapped together which is at least better than nothing."
Stick that in front of your wall of LLM text, and I'd at least have confidence that some day you're going to come back and make the README readable.
IMO we need less "photoshop and blender for AI", more `rails new my_proj` for video.
I have a project like this: it setups up a scaffolding: scripts, audio tracks, hot-reloading authoring UI, TTS/STT, sfx, music, shared-styles, beats, video composing, time-control for deterministic rendering, etc. Then you can use any web-technology to render (webGPU, Lottie, shadcn, SVG, whatever).
I've been doing little informational videos with Opus 5.5 [1]. The jump in capability floored me the first time I saw it in action. Clearly something happened in this last generation. That particular video took it about 45 minutes and I think it spent less than a $1 on OpenRouter for the voice over / sound.
The readme is too busy, but the video showcase is impressive [0].
I agree the framing is too arrogant though, about replacing blender etc, except if it's delegating the computing to those, because it is very complicated.
Definitely using it for my next promo-site which is for event that comes in one month. People be blown. Also I wonder what was used for the audio synthesis, if author reads.
here's my last work that need quite a lot of work to get where it is, but I know what I do also:
I am clearly not the target audience, but as a hobbyist that has used da vinci resolve at points i understand basically nothing of what the software actually is supposed to do
basically it is a toolkit for agents (or developers) to create and render media using code. it's video editing where you describe the video instead of clicking or dragging stuff.
I think YouTube just defaults to a lower resolution (was like 360p in auto when I opened it) maybe because it's mostly text (cause I noticed that too), but you can change it to be 1080p on the settings cog on the video.
Oh, brilliant. I can cut video with a single click and rearrange clips by dragging and dropping, but now I will have to install npm (and all of the vulnerabilities) to dick around with LLMs via a bunch of APIs.
Oh, brilliant. I can do source control by just sending .diff file via email to my colleges...
No but to be serious, I don't know how well this works but it CLEARLY targets people who want motion graphic animation / keyframe stuff, that you normally would do with AE. If you just want to cut a few clips it's not even targeting you. No reason to get all upset about it.
Look, you can’t just write this. When you say something like this, you set the bar extremely high, and open yourself up to (totally fair!) criticism if Photoshop or After Effects gurus point out entire classes of use cases you aren’t handling.
If you want people to talk about what your tool is good at, you need to stop inviting comparisons to titans like this. It will take you a long long time to build something as fully featured as them, and there is plenty of room for agent-focused tools that aren’t at that immense product scale.
It's cool that so many people are building useful tools with LLMs, but for the love of ones and zeroes, people! PLEASE write your README by hand, instead of handing it off to the LLM with no human involvement! You know how unreadable LLM prose is! Why do you make that the first thing people see when they see your repo?
Stick that in front of your wall of LLM text, and I'd at least have confidence that some day you're going to come back and make the README readable.
I have a project like this: it setups up a scaffolding: scripts, audio tracks, hot-reloading authoring UI, TTS/STT, sfx, music, shared-styles, beats, video composing, time-control for deterministic rendering, etc. Then you can use any web-technology to render (webGPU, Lottie, shadcn, SVG, whatever).
https://github.com/scosman/videowright
[1]: https://www.youtube.com/watch?v=ijFFpMpS8W0
I agree the framing is too arrogant though, about replacing blender etc, except if it's delegating the computing to those, because it is very complicated.
[0] https://www.youtube.com/watch?v=w6IQNrhJJek
Someone at work recorded a demo of a new feature using Claude + https://github.com/veedstudio/open-edit
It was pretty slick and he said he only did two passes on it. Not sure what it means for Adobe, but hopefully it lights a fire under them.
here's my last work that need quite a lot of work to get where it is, but I know what I do also:
bsf.hmsu.org
works on mobile as old as iphone12
Wouldn't that be trivially easy to do?
No but to be serious, I don't know how well this works but it CLEARLY targets people who want motion graphic animation / keyframe stuff, that you normally would do with AE. If you just want to cut a few clips it's not even targeting you. No reason to get all upset about it.