epicenv 1.1.0

Schema-based environment variable management for Python projects

epicenv

Schema-based environment variable management for Python projects.

Define your environment variables in pyproject.toml with types, defaults, and help text. Use epicenv to create, validate, and manage .env files. Works with any Python project - Django, FastAPI, Flask, or plain Python.

Features

• 📋 Schema-based: Define variables in pyproject.toml with types, defaults, and documentation
• ✅ Validation: Catch undefined variables in debug mode before they cause runtime errors
• 🛠️ CLI Tools: Create and compare .env files with uvx epicenv create and uvx epicenv diff
• 🎯 Framework Agnostic: Works with Django, FastAPI, Flask, or any Python project
• 🔧 Flexible: Use schema-only, code-only, or hybrid approaches
• 🐍 Type Safe: Full type support (str, bool, int, list, url, json, etc.)

Quick Start

1. Define your schema in pyproject.toml

[tool.epicenv.variables]
SECRET_KEY = {
    type = "str",
    required = true,
    help_text = "Secret key for cryptographic signing",
    initial_func = "secrets.token_urlsafe"
}

DEBUG = {
    type = "bool",
    default = false,
    initial = "on",
    help_text = "Enable debug mode (never enable in production)"
}

DATABASE_URL = {
    type = "dj_db_url",
    default = "sqlite:///db.sqlite3",
    help_text = "Database connection URL"
}

API_KEY = {
    type = "str",
    required = true,
    help_text = "API key for external service"
}

2. Create a .env file

Use uvx to run without installing:

uvx epicenv create

Or install and use directly:

uv add epicenv
epicenv create

This generates a .env file with help text, types, and initial values:

# This is an initial .env file generated on 2026-01-24...
# Any environment variable with a default can be safely removed or commented out.

# Secret key for cryptographic signing
# type: str
SECRET_KEY=hkXzH0ZFG1YOtKhDsXAqS7rkt50_fZu_bV8YzPXr7VA

# Enable debug mode (never enable in production)
# type: bool
# default: False
DEBUG=on

# Database connection URL
# type: dj_db_url
# default: sqlite:///db.sqlite3
# DATABASE_URL=

# API key for external service
# type: str
API_KEY=

3. Use in your code

from epicenv import Env

env = Env()
env.read_env()  # Load from .env file

SECRET_KEY = env.str("SECRET_KEY")
DEBUG = env.bool("DEBUG", default=False)
DATABASE_URL = env.str("DATABASE_URL", default="sqlite:///db.sqlite3")
API_KEY = env.str("API_KEY")

4. Validate your environment

Check for missing required variables:

epicenv validate

Compare your .env file with the schema:

epicenv diff

Installation

For any Python project

uv add epicenv

For Django projects (with dj-database-url, dj-email-url support)

uv add epicenv[django]

Without installation (using uvx)

uvx epicenv create
uvx epicenv diff
uvx epicenv validate

CLI Commands

epicenv create

Create a .env file from your pyproject.toml schema.

epicenv create                    # Create .env in current directory
epicenv create --path config/.env # Create at specific path
epicenv create --no-backup        # Don't backup existing file

epicenv diff

Compare your .env file with the schema and show differences.

epicenv diff                    # Compare .env with schema
epicenv diff --path config/.env # Compare specific file

Output:

✓ All variables are in sync!

or

Missing required variables:
  • API_KEY - API key for external service

Missing optional variables (have defaults):
  • DATABASE_URL (default: sqlite:///db.sqlite3) - Database connection URL

Variables in .env file not defined in schema:
  • OLD_VARIABLE

epicenv validate

Validate current environment against schema.

epicenv validate          # Validate environment
epicenv validate --strict # Exit with error code if validation fails

Validation Mode

Validation is automatically enabled when DEBUG=true. Control validation behavior with the EPICENV_VALIDATE environment variable:

• auto (default): Validate when DEBUG=true, otherwise no validation
• strict: Always validate and raise errors for undefined variables
• warn: Always validate but only warn (don't raise errors)
• off: Disable validation completely

# Auto mode (default) - validates only when DEBUG=true
DEBUG=true python manage.py runserver  # Validates!
DEBUG=false python manage.py runserver # No validation

# Force strict validation regardless of DEBUG
EPICENV_VALIDATE=strict python manage.py runserver

# Warn about undefined variables but don't fail
EPICENV_VALIDATE=warn python manage.py runserver

# Disable validation completely
EPICENV_VALIDATE=off python manage.py runserver

In your code:

from epicenv import Env

env = Env()  # Validation mode controlled by EPICENV_VALIDATE env var
env.read_env()

# If MY_VAR is not in pyproject.toml schema and validation is enabled:
value = env.str("MY_VAR")  # UndefinedVariableError!

Error message:

UndefinedVariableError: Environment variable 'MY_VAR' is not defined in pyproject.toml schema.

Add it to [tool.epicenv.variables] in your pyproject.toml:

[tool.epicenv.variables]
MY_VAR = { type = "str", help_text = "Description here" }

Or disable validation by setting EPICENV_VALIDATE=off

Schema Reference

Field Types

All environs types are supported:

• Basic: str, bool, int, float, decimal
• Collections: list, dict, json
• Date/Time: date, datetime, time, timedelta
• Specialized: url, uuid, path, enum, log_level
• Django (with [django] extra): dj_db_url, dj_email_url, dj_cache_url

Schema Fields

[tool.epicenv.variables]
MY_VARIABLE = {
    type = "str",              # Variable type (required)
    required = true,           # Is this variable required? (default: true if no default)
    default = "value",         # Default value for .env generation and diff
    help_text = "Description", # Documentation for this variable
    initial = "initial_value", # Static initial value for .env generation
    initial_func = "module.function"  # Callable for dynamic initial values
}

Understanding default vs Runtime Defaults

The default field in pyproject.toml is only used for .env file generation and diff commands. It is not used at runtime.

Runtime defaults must always be specified in your Python code:

# Runtime default is specified here, not in pyproject.toml
DEBUG = env.bool("DEBUG", default=False)

# This allows for dynamic defaults based on other values
SECURE_COOKIES = env.bool("SECURE_COOKIES", default=not DEBUG)

This design prevents confusion about which default takes precedence and allows for dynamic defaults that depend on other values.

Initial Value Functions

Use initial_func to generate dynamic values:

[tool.epicenv.variables]
SECRET_KEY = {
    type = "str",
    initial_func = "secrets.token_urlsafe"  # Python stdlib
}

DJANGO_SECRET = {
    type = "str",
    initial_func = "django.core.management.utils.get_random_secret_key"
}

CUSTOM_VALUE = {
    type = "str",
    initial_func = "myapp.utils.generate_api_key"  # Your own function
}

Built-in Initializers

epicenv includes built-in initializer functions in the epicenv.initializers module:

epicenv.initializers.url_safe_password - Generate a URL-safe random password (letters, digits, -, _):

[tool.epicenv.variables]
SECRET_KEY = {
    type = "str",
    required = true,
    help_text = "Secret key for cryptographic signing",
    initial_func = "epicenv.initializers.url_safe_password"
}

# With custom length (default is 50)
API_TOKEN = {
    type = "str",
    required = true,
    help_text = "API authentication token",
    initial_func = "epicenv.initializers.url_safe_password",
    kwargs = { length = 32 }
}

More initializers may be added in the future, such as fetching secrets from 1Password.

Custom Initializers

You can create your own initializer functions anywhere in your codebase:

# myapp/env_utils.py
import secrets

def generate_api_key() -> str:
    """Generate an API key with a custom prefix."""
    return f"sk_{secrets.token_hex(16)}"

def generate_uuid() -> str:
    """Generate a UUID string."""
    import uuid
    return str(uuid.uuid4())

Then reference them in your schema:

[tool.epicenv.variables]
API_KEY = {
    type = "str",
    required = true,
    initial_func = "myapp.env_utils.generate_api_key"
}

INSTANCE_ID = {
    type = "str",
    required = true,
    initial_func = "myapp.env_utils.generate_uuid"
}

Django Integration

Option 1: Schema-based (Recommended)

Define variables in pyproject.toml and use the Env class:

# settings.py
from pathlib import Path
from epicenv import Env

BASE_DIR = Path(__file__).resolve().parent.parent

env = Env()  # Validation automatically enabled when DEBUG=true
env.read_env(BASE_DIR / ".env")

SECRET_KEY = env.str("SECRET_KEY")
DEBUG = env.bool("DEBUG", default=False)
DATABASES = {"default": env.dj_db_url("DATABASE_URL", default="sqlite:///db.sqlite3")}

Then use CLI commands:

epicenv create  # Instead of ./manage.py create_env_file
epicenv diff    # Instead of ./manage.py diff_env_file

Option 2: Django Management Commands (Legacy)

For backward compatibility, Django management commands still work:

# settings.py
INSTALLED_APPS = [
    ...
    "epicenv",  # Add to INSTALLED_APPS
]

from epicenv import Env

env = Env()
env.read_env(BASE_DIR / ".env")

# Use help_text, initial, initial_func as before
SECRET_KEY = env.str(
    "SECRET_KEY",
    initial_func="django.core.management.utils.get_random_secret_key",
    help_text="Django's secret key",
)

Then use Django commands:

./manage.py create_env_file
./manage.py diff_env_file

Examples

FastAPI Project

# pyproject.toml
[tool.epicenv.variables]
APP_NAME = { type = "str", default = "My API", help_text = "Application name" }
API_HOST = { type = "str", default = "0.0.0.0", help_text = "API host" }
API_PORT = { type = "int", default = 8000, help_text = "API port" }
DATABASE_URL = { type = "url", required = true, help_text = "Database URL" }
REDIS_URL = { type = "url", default = "redis://localhost:6379", help_text = "Redis URL" }
LOG_LEVEL = { type = "log_level", default = "INFO", help_text = "Logging level" }

# config.py
from epicenv import Env

env = Env()
env.read_env()

APP_NAME = env.str("APP_NAME", default="My API")
API_HOST = env.str("API_HOST", default="0.0.0.0")
API_PORT = env.int("API_PORT", default=8000)
DATABASE_URL = env.url("DATABASE_URL")
REDIS_URL = env.url("REDIS_URL", default="redis://localhost:6379")
LOG_LEVEL = env.log_level("LOG_LEVEL", default="INFO")

Flask Project

# pyproject.toml
[tool.epicenv.variables]
FLASK_APP = { type = "str", default = "app.py", help_text = "Flask app entry point" }
FLASK_ENV = { type = "str", default = "production", help_text = "Flask environment" }
SECRET_KEY = { type = "str", required = true, help_text = "Flask secret key", initial_func = "secrets.token_hex" }
DATABASE_URL = { type = "url", required = true, help_text = "Database URL" }

# config.py
from epicenv import Env

env = Env()
env.read_env()

class Config:
    FLASK_APP = env.str("FLASK_APP", default="app.py")
    FLASK_ENV = env.str("FLASK_ENV", default="production")
    SECRET_KEY = env.str("SECRET_KEY")
    SQLALCHEMY_DATABASE_URI = env.url("DATABASE_URL")

Contributing

Contributions are welcome! Please open an issue or submit a pull request.

License

This project is licensed under the MIT License.