Zero-config project setup

Most projects come with a README that’s half right, a .env.example that’s half full, and a setup script that worked on the author’s laptop in 2023. You spend an afternoon getting it to boot before you write your first line of code.

Loopyard skips that afternoon. Point it at a git repo and the setup agent reads the codebase the way a new hire would: package.json, Gemfile, mix.exs, requirements.txt, the migration folder, the docker-compose file if there is one. It figures out the language, the framework, the databases, the queue, and what your dev server is called. Then it writes the Dockerfile, writes the compose file, fills in sensible defaults, and starts everything up.

You watch it work in real time. The terminal stream is live, the chat is live, the service logs are live. If it gets stuck on something genuinely project-specific, you tell it; it doesn’t ask you the same question twice.

By the time you’d have finished reading the README, the dev server is up, the database is migrated, and an agent is sitting in the workspace container waiting for the first task.