backendctl

Stop rewriting the same boilerplate. Start with a solid foundation.

backendctl is an interactive CLI that generates a batteries-included backend project — auth, database, migrations, rate limiting, tests, linting, and optional AI-assistant config — for the framework of your choice. Pick FastAPI, Flask, or Django REST Framework, answer a few questions, and start coding your actual business logic.

Features

• Three frameworks — FastAPI (SQLModel), Flask (SQLAlchemy), or Django REST Framework.
• JWT auth out of the box — register / login / refresh, Argon2 password hashing, a User model, and a /users/me endpoint.
• Database choice — PostgreSQL (with SQLite for tests), MongoDB, or both.
• Sane defaults — rate limiting, CORS, env-based config, layered settings.
• Migrations — Alembic (FastAPI), Flask-Migrate, or Django migrations.
• Tests included — pytest suite wired up with fixtures.
• Tooling — ruff, mypy, .pre-commit-config.yaml, .editorconfig.
• AI-assistant ready — optional CLAUDE.md/AGENTS.md, .cursorrules, mcp.json, and SDK dependency.
• uv or pip — your choice of package manager.

Installation

Requires Python 3.11+.

# Run without installing (recommended)
uvx backendctl new

# Or install as a tool
uv tool install backendctl

# Or with pip
pip install backendctl

From source

git clone https://github.com/dipto0321/backendctl
cd backendctl
uv sync
uv run backendctl --help

Usage

Interactive wizard

backendctl new

You'll be guided through:

1. Project name
2. Package manager (uv / pip)
3. Framework (FastAPI / Flask / Django REST Framework)
4. Database (PostgreSQL+SQLite / MongoDB / both)
5. Database name / user / password (defaults: slug / slug / auto-generated)
6. Authentication (JWT / none)
7. Optional name field on the User model
8. AI-assistant config (Claude / OpenAI / none)
9. Git initialization

Non-interactive flags

Any flag you pass skips the matching wizard step:

backendctl new myapi --framework fastapi --db postgres --pm uv --no-ai

What you get

After generation, backendctl runs pre-flight checks, writes the project, sets up git (optional), and installs dependencies. A docker-compose.yml is generated for your chosen database, and .env ships with working credentials that already match it — no manual copying or editing required to get a database running. Then:

cd myapi
docker compose up -d      # start the database
uv run fastapi dev        # or: flask run / python manage.py runserver

Generated project layout

myapi/
├── src/myapi/
│   ├── main.py                 # app factory + lifespan
│   ├── core/                   # config, database, security
│   ├── middleware/             # rate limiting
│   ├── api/v1/router.py        # versioned router
│   └── modules/
│       ├── auth/               # models, schemas, service, deps, router
│       └── users/              # /me endpoint
├── alembic/                    # migrations
├── tests/                      # pytest suite
├── pyproject.toml
└── .env.example

myapi/
├── src/myapi/
│   ├── __init__.py             # app factory
│   ├── config.py               # Config / TestConfig
│   ├── extensions.py           # db, jwt, limiter, migrate
│   ├── models/user.py
│   └── blueprints/
│       ├── auth/               # routes + service
│       └── users/              # /me endpoint
├── migrations/                 # Flask-Migrate
├── tests/
├── pyproject.toml
└── .env.example

myapi/
├── config/
│   ├── settings/               # base / development / production
│   ├── urls.py, wsgi.py, asgi.py
├── apps/
│   ├── users/                  # custom User model + /me
│   └── authentication/         # register + JWT token views
├── core/                       # pagination, exception handler
├── tests/
├── manage.py
├── pyproject.toml
└── .env.example

Development

uv sync                 # install dev dependencies
uv run pytest           # run the test suite
uv run ruff check src   # lint
uv run ruff format src  # format
uv run mypy src         # type-check

The repository is laid out as:

src/backendctl/
├── main.py              # Typer entry point
├── cli/new.py           # `new` command + interactive wizard
├── core/                # config dataclasses, console helpers, pre-flight checks
├── generators/          # per-framework file writers (BaseGenerator + subclasses)
└── templates/           # pure functions returning file contents

To add or change generated output, edit the relevant function in templates/ and the matching _scaffold() in generators/. The test suite compiles every generated .py file, so syntax regressions are caught automatically.

License

MIT © Dipto Karmakar