Skip to content

Moteo

Video composition application using GPT-5 as a backbone to prompt and create motion graphics from code.

Core trick

GPT-5 writes Remotion React code from a text prompt, and a live Remotion Player renders it as a preview.

Tags

Description

Video composition application using GPT-5 as a backbone to prompt and create motion graphics from code. Users prompt the visuals they want in natural language, retool the result and compose them together into larger projects. Each video can be converted into a "block" which is automatically parameterized and can be added as pieces to other videos, which themselves can be blocked as well. The LLM takes in visuals, textual data, research context as parameters to each video block to create graphics, animations, etc. anchored to data and contextually linked via the LLM. The result is LLM as a context-manager for a graphics studio that can export to apps, chain data, and be automated to generate based on fed parameters which semantically evolve and develop with the content of the video.
Sami1309/snackathon★ 00 forksTypeScriptCreated Aug 10, 2025 · last push Aug 10, 2025

README excerpt

# Prompt → Remotion (One‑Page App + Backend)

This repo includes:
- A backend (`server.js`) that proxies prompts to OpenAI (defaults to `gpt-5`) and returns a Remotion project (TypeScript/React).
- A single‑page app in `web/` with a live Remotion Player that previews the generated code and shows a progress bar.

If you opened a URL and saw nothing, make sure you are using the Vite dev server URL during development: `http://localhost:5173`.

## Quick Start (Dev)

- Prerequisites: Node.js 18+
- Set key (real API): `export OPENAI_API_KEY=sk-...`
- Start backend on port 3003:
  - `PORT=3003 node server.js`
- Start the web app (Vite dev server on 5173):
  - `cd web && npm install && npm run dev`
- Open the app: `http://localhost:5173`
- In the app, uncheck “Use mock backend”, write a prompt, click “Generate Remotion Project”. The right panel previews it live.

Troubleshooting blank page
- Open the correct URL: `http://localhost:5173` (dev server). The Node server at `http://localhost:3003` only serves the built app (`web/dist`) after you run `npm run build`.
- Check backend health: `curl http://localhost:3003/api/health` — ensure `hasKey: true` when using the real API.
- Network restrictions: If outbound calls are blocked, use mock mode in the UI, or run `DEV_FALLBACK=1 PORT=3003 node server.js` so the backend returns a local example.
- Logs: `tail -f server.log` and `tail -f web/vite.log` if you started them in the background.

## Production (Serve the built SPA from Node)

- Bui

Team