discovery + beautiful reading UX for papers feels overdue. Curious to see how the on-device AI and quiz mechanics actually change the way people explore and retain new research.
Cool idea – the local LLM space really needs a tool like this that actually understands hardware and real benchmarks instead of just “biggest model that fits.”
Didn’t expect a few lines of APL plus ballot-sequence magic to make random plane trees feel this intuitive—super elegant construction. Curious how this could be applied in practice, e.g. generative graphics or random UI/tree structures?
like how thin this harness is. Letting the LLM work directly against CDP with just helpers.py + SKILL.md feels much closer to how I want “skills” to look: small, inspectable, and self-extending when the agent actually needs something (like that upload_file moment). Curious to see how far this pattern can go beyond the browser.
This looks great. Curious about the lip-sync — viseme set or just
open/closed mouths? The South Park style is super forgiving but
HyperFrames quality seems like it'd need more.