Price terms as published
Each line is the vendor’s own wording, split from the price capture checked on 2026-10-27. Nothing here is rounded, converted or re-worded — the full sentences stay in the review below. Lines that carry only a price continue the plan named on the line above, in the vendor’s own order.
| # | Plan / term | Checked |
|---|---|---|
| 1 | Free — MIT licence for the code, CC BY 4.0 for the style prompts and films | 2026-10-27 |
| 2 | no paid tier, no hosted service | 2026-10-27 |
One-liner: Lemo-Opuscar is a collection of 59 film styles — each one a prompt plus a short film made entirely in code by Claude Opus 5.5 — designed to be picked up and directed by your coding agent, with no video model in the pipeline.
This entry is compiled from the project’s GitHub repository and README, read on 2026-10-27. It is not a hands-on test: no film was generated for this page, no score is assigned, and every number below is sourced inline.
Verified facts (GitHub API and README, captured 2026-10-27)
| Attribute | Value |
|---|---|
| Repository | lemomo-ai/lemo-opuscar |
| Stars | 8,414 (GitHub API, 2026-10-27) |
| Created / last push | 2026-09-26 / 2026-10-24 |
| Licence | MIT (code) + CC BY 4.0 (style prompts and films, per README badge) |
| Price | $0 — no paid tier, no hosted service mentioned anywhere in the README |
| Ready-made styles | 59, each with a style prompt and a demo film |
| Homepage / gallery | https://lemomo-ai.github.io/lemo-opuscar/ |
| README languages | English + 简体中文 |
| Video model involved | None — every film is code-rendered (the README says so explicitly) |
The star count was pulled directly from the GitHub API on 2026-10-27; the licence and style-count claims come from the README badge and body, which also mark the project as “Built for Claude.”
What it is
The repository answers one workflow question: “what should I tell my coding agent to make a film look like?” Each of the 59 styles bundles the recipe — the style prompt — with a finished demo film that was made entirely by Claude Opus 5.5 writing code. The README’s framing is: pick a style or make one from scratch, bring your story, and let your coding agent direct.
Because the demos ship with their prompts and source, a style is not a black box: you can watch the reference film, read the prompt that produced it, and reuse the same language with your own story. The project also explains how to build your own style from scratch, which turns the collection into a kit rather than a gallery.
How it works
- Each style = a style prompt (what to tell the agent) + a demo film (what “done” looks like).
- Films are rendered as code — HTML/Canvas/SVG output — so they are deterministic and viewable in any browser.
- No video model is used: the README is explicit that the entire film is generated by a coding agent writing and rendering code.
- The gallery plays the 59 reference films directly, and every style page in the repo keeps its prompt next to its source, so the recipe is readable even without watching the film.
Because every style is self-contained, the collection scales in both directions: beginners can copy a finished style verbatim, while people building their own look can read how existing styles were prompted and start from a proven pattern instead of a blank page. The README is bilingual (English and 简体中文), which is a deliberate signal that the kit is aimed at a global audience of Claude Code users rather than a single community.
Install & run
- Open the gallery at
https://lemomo-ai.github.io/lemo-opuscar/and pick a style (or readREADME.md/README.zh-CN.md). - Copy the style’s prompt from the repo.
- Run it in Claude Code (or any compatible coding agent) with your story text.
- Optionally follow the “make your own style” section in the README to contribute a new one.
Price & license
Free. The code is MIT-licensed and the style prompts and films are CC BY 4.0, per the README badges. There is no paid tier and no hosted service. The real cost is your agent’s compute time — films are generated frame-by-frame as code, which takes minutes rather than the hours of GPU rendering a diffusion model would use.
Alternatives
- mg-styles-15 — the same idea with 15 motion-design styles and complete source per film; smaller, easier to audit.
- awesome-opus5-5-videos — 703 community prompts behind viral Opus 5.5 videos; broader but less curated (its README links to the commercial Skillry site).
- pdoom-video — a single, deep example: a full code-rendered music video project with its treatment and engine documented.
- Video-model generators (Sora, Veo, Runway, Kling) — a different approach: they render with diffusion models, so you get photorealistic footage rather than deterministic code output; they are also paid services.