deploy django

the whole thing

curl -fsSL https://pushto.host/install | sh   # or: brew install mydude-industries/tap/pushto · npm i -g @pushto/cli
pushto login
pushto init my-django
pushto

that last bare pushto is the deploy. on windows, the install line is irm https://pushto.host/install.ps1 | iex in powershell. live at https://<your-name>.pushto.host with https. we build it and run it as a server. you never touch the machinery. runs on the free tier too — your one free app can be a server; it naps after a quiet spell and wakes on the next visit. $13/mo lifts the one-project cap and keeps apps always awake.

how we know it's django

• django listed in requirements.txt, Pipfile, or pyproject.toml at the project root

what happens when you push

• gunicorn serves <your project>.wsgi:application on port 8080. the project name comes from where your wsgi.py lives
• python manage.py migrate runs automatically before the server starts. a failed migration does not block startup

sharp edges

• more than one wsgi.py and the guess stops: the fallback target is config.wsgi:application. keep a single wsgi.py, or name your project folder config
• a sibling frontend/package.json with a build script gets built first, and its dist folder is there for django to serve

deploying from an agent?

claude code and cursor can run all of this for you: connect your agent. the short machine-readable version of this site lives at https://pushto.host/llms.txt.