Skip to main content
Neptune provides managed Postgres databases for storing application data.

​
Quick Start

Ask your AI assistant:
“Add a Database called ‘details’ to my project”
Or add it directly to your neptune.json:
neptune.json
{
  "kind": "Service",
  "name": "my-app",
  "resources": [
    {
      "kind": "Database",
      "name": "details"
    }
  ]
}
Then ask your AI assistant to provision and deploy:
“Provision my resources and deploy”

​
How It Works

When you provision a database, Neptune:
  1. Creates a managed Postgres database
  2. Stores the connection string securely as a secret
  3. Exposes the connection string to your app as an environment variable
  4. You update your application to read that environment variable (and redeploy if necessary)
For the time being, you cannot connect to the database directly from your local machine. We will be releasing tooling that lets you query your database from your AI assistant.

​
Using in Your Application

The connection string is available as an environment variable: DATABASE_URL_{RESOURCE_NAME}. For example, if your database resource is named details, you’ll get DATABASE_URL_DETAILS. If you name your database my-app-db, you’ll get DATABASE_URL_MY_APP_DB. If you want to use a different name for your database, you can specify it in your neptune.json: 
{
  "kind": "Database",
  "name": "my-app-db"
}

import os
import psycopg2

# Connection string is automatically available as environment variable
conn = psycopg2.connect(os.environ['DATABASE_URL_DETAILS'])
cursor = conn.cursor()
cursor.execute("SELECT version();")
print(cursor.fetchone())
conn.close()
Your AI assistant can update your code to read the correct DATABASE_URL_... variable and add the right Postgres driver for your language/framework.

​
Configuration Options

{
  "kind": "Database",
  "name": "details"
}
FieldTypeRequiredDescription
kindstringYesMust be "Database"
namestringYesName of the database.