The Zyeta backend uses environment variables for configuration. This guide explains how to set up these variables for development and production environments.

​
Environment File (.env)

Create a .env file by copying the provided .env.local template: 
cp .env.local .env
The .env file contains sensitive information and should never be committed to version control. Note that it’s already included in the .gitignore file.

​
Required Environment Variables

Based on the application’s settings module, the following environment variables are required: 
# Application
APP_NAME=zyeta.backend
ENVIRONMENT=dev  # Can be dev | beta | prod

# Security
JWT_SECRET=your_jwt_secret_key
MASTER_API_KEY=your_master_api_key
JWT_EXPIRE_MINUTES=1400  # Default: 24 hours

# Database
DATABASE_URL=postgresql+asyncpg://postgres:mysecretpassword@localhost:5432/postgres

# External Services
OPENAI_API_KEY=your_openai_api_key
ANTHROPIC_API_KEY=your_anthropic_api_key
RESEND_API_KEY=your_resend_api_key
FIRECRAWL_API_KEY=your_firecrawl_api_key

# Frontend
FRONTEND_URL=http://localhost:8000

# Storage
S3_BUCKET=zyeta-dev
S3_ACCESS_KEY=your_s3_access_key
S3_SECRET_KEY=your_s3_secret_key
S3_ENDPOINT=https://s3.backendly.io
PUBLIC_S3_BUCKET=public-assets

# Payment
STRIPE_SECRET_KEY=your_stripe_secret_key
STRIPE_PUBLISHABLE_KEY=your_stripe_publishable_key
STRIPE_WEBHOOK_SECRET=your_stripe_webhook_secret

# Task Queue
CELERY_BROKER_URL=redis://localhost:6379/0
CELERY_RESULT_BACKEND=redis://localhost:6379/0

# Knowledge Base
KB_SETTINGS_VERSION=1

# Sandbox
PYTHON_SANDBOX_TESTING_URL=http://localhost:3000

​
Environment Variable Details

​
Application Settings

  • APP_NAME: Name of your application
  • ENVIRONMENT: Current environment (dev, beta, prod)

​
Security

  • JWT_SECRET: Secret key for signing JWT tokens
  • MASTER_API_KEY: Master API key for administrative access
  • JWT_EXPIRE_MINUTES: JWT token expiration time in minutes

​
Database

  • DATABASE_URL: Connection string for PostgreSQL database Format: postgresql+asyncpg://user:password@host:port/database Examples:
    • Local Docker: postgresql+asyncpg://postgres:mysecretpassword@localhost:5432/postgres
    • Supabase: postgresql+asyncpg://postgres:password@db.izufeytmzeanvqoyckwp.supabase.co:5432/postgres

​
External Services

  • OPENAI_API_KEY: API key for OpenAI services
  • ANTHROPIC_API_KEY: API key for Anthropic AI services
  • RESEND_API_KEY: API key for Resend email service
  • FIRECRAWL_API_KEY: API key for Firecrawl service

​
Frontend

  • FRONTEND_URL: URL of the frontend application

​
Storage

  • S3_BUCKET: Name of the S3 bucket for storage
  • S3_ACCESS_KEY: Access key for S3
  • S3_SECRET_KEY: Secret key for S3
  • S3_ENDPOINT: Endpoint URL for S3 service
  • PUBLIC_S3_BUCKET: Name of the public S3 bucket

​
Payment

  • STRIPE_SECRET_KEY: Secret key for Stripe
  • STRIPE_PUBLISHABLE_KEY: Publishable key for Stripe
  • STRIPE_WEBHOOK_SECRET: Webhook secret for Stripe

​
Task Queue

  • CELERY_BROKER_URL: URL for Celery broker (Redis)
  • CELERY_RESULT_BACKEND: URL for Celery result backend (Redis)

​
Knowledge Base

  • KB_SETTINGS_VERSION: Version of knowledge base settings

​
Sandbox

  • PYTHON_SANDBOX_TESTING_URL: URL for Python sandbox testing

​
Alembic Configuration

The project includes an alembic.copy.ini template that you should copy to create your own alembic.ini: 
cp alembic.copy.ini alembic.ini
Then update the sqlalchemy.url field in alembic.ini to match your database connection string. Note that you need to escape percent signs (%) with another percent sign (%%): 
# For local Docker PostgreSQL
sqlalchemy.url = postgresql+asyncpg://postgres:mysecretpassword@localhost:5432/postgres

# For Supabase (with percent sign in password)
sqlalchemy.url = postgresql+asyncpg://postgres:Test%%401234@db.eewruhvfyvvvlmlpvphw.supabase.co:5432/postgres

​
Environment-Specific Configuration

​
Development Environment

For development, you can use local services: 
ENVIRONMENT=dev
DATABASE_URL=postgresql+asyncpg://postgres:mysecretpassword@localhost:5432/postgres
CELERY_BROKER_URL=redis://localhost:6379/0
CELERY_RESULT_BACKEND=redis://localhost:6379/0
PYTHON_SANDBOX_TESTING_URL=http://localhost:3000

​
Testing Environment

For testing, use separate databases and services: 
ENVIRONMENT=test
DATABASE_URL=postgresql+asyncpg://postgres:mysecretpassword@localhost:5432/zyeta_test

​
Production Environment

For production, use secure, production-grade services: 
ENVIRONMENT=prod
DATABASE_URL=postgresql+asyncpg://postgres:password@production-db-host:5432/zyeta
PYTHON_SANDBOX_TESTING_URL=https://python-sandbox-testing.backendly.io

​
Loading Environment Variables

The application uses Python’s dotenv package to load environment variables from the .env file. The settings are loaded using Pydantic’s BaseSettings class, which provides type validation and default values. 
from dotenv import load_dotenv
from pydantic_settings import BaseSettings

load_dotenv()

class Settings(BaseSettings):
    # Settings defined here
    
    class Config:
        env_file = ".env"
        env_file_encoding = "utf-8"

​
Accessing Settings in Code

The application uses a singleton pattern for settings. Import the settings object to access environment variables: 
from config.settings import settings

# Example usage
database_url = settings.database_url
jwt_secret = settings.jwt_secret

​
Keeping Secrets Secure

Never commit sensitive information like API keys or database passwords to version control. Always use environment variables or secure secret management solutions.

​
Recommendations for Secret Management

  1. Development: Use local .env files
  2. CI/CD: Use your CI/CD provider’s secret management
  3. Production: Use a secret manager like AWS Secrets Manager or HashiCorp Vault

​
Troubleshooting