Open-sourced 39 video styles drawn entirely by Opus 5.5 code, no video model or stock footage
EricBuess · x · 2026-09-27
lemoai open-sourced Lemo-opuscar: after making 100+ videos with Claude Opus 5.5, they curated 39 covering styles from ink wash to shadow puppets, cel animation, pixel RPG, and pop-up books—every frame rendered frame-by-frame by code the model wrote (Canvas/WebGL), with original music from free sample libraries, local TTS voiceover, and self-mixing/review. No video generation model or stock footage involved.
Key lessons: stable output depends less on fancy prompts than on (1) giving the agent a mature methodology and (2) stating requirements precisely. The repo ships three docs: DIRECTOR.md (story, pacing, shots, performance, self-check), TECHNIQUE.md (frame rendering, VO, music, mixing, review), and per-style STYLE.md files. Usage: clone the repo, launch claude, and ask for e.g. 'a 45-second watercolor-style film about my family's coffee farm'—the agent reads the docs and produces the video.
More from coding & agent
- DHH on AI coding: thinking "architecture" is your fortress? Models are good at that too — tawnniee · 2026-09-27
- Reverse engineering with coding agents: set it up overnight, wake up to source — moyix · 2026-09-27
- ThePrimeagen joins DHH's Omarchy Core to lead agentic QA with Oligarchy harness — mark_k · 2026-09-27
- LLMs don't write better code — they accelerate the glue work of integrating libraries — markjeffrey · 2026-09-27
- Open-source AI agents now delegate tasks to each other: GrokBot + Muse — Saboo_Shubham_ · 2026-09-27
- 'Agentic AI Killed DevOps' Claim Sparks Developer Pushback — emeka_boris · 2026-09-27