CubeDesk is a free, open-source app for speedcubers. Track your times, see your stats, train hundreds of algorithms, and race other cubers 1v1.
Come say hi on Discord. It's the best place for questions, feedback, and feature ideas.
Found a bug? Open an issue with the bug label.
Contributions are welcome, thank you for helping out!
- For anything bigger than a small fix, check in on Discord or open an issue first so we can agree on the approach.
- Keep pull requests focused on one change.
- Run the app and test your change yourself before opening a PR, and make sure you understand every line you're submitting.
You'll need Node 24 and either Docker or Postgres and Redis already running. On Linux, the setup script installs everything for you.
git clone https://github.com/kash/cubedesk.git
cd cubedesk
bash scripts/dev-setup.sh
pnpm devOpen localhost:3000 and log in as agent@cubedesk.test with the password cubedesk. The setup script installs dependencies, creates .env, sets up the database, and seeds it with demo accounts, solves, friends, and 1v1 history.
To add about 4,000 sample solves to an account you created locally:
pnpm seed:dev --username YOUR_USERNAMETo reset the database and start over:
pnpm exec prisma migrate reset --force && pnpm seed:devThe seed script only runs against a local development database.
Pushing to main deploys to production on ECS Fargate through GitHub Actions. pnpm build builds the client into dist/ and the server into build/server/, and pnpm start runs the built server. Production environment variables live in the ECS task definition. The SENTRY_AUTH_TOKEN secret uploads source maps during the build, and SENTRY_DSN turns on backend error reporting at runtime.
CubeDesk is licensed under the GPL v3 or later. You're free to use it, but anything you build on it must also be open source.
I chose this license so the cubing community can build on each other's work.