Skip to main content

Deploy with an agent

Copy agent prompt

Before you start

Connect your AWS account and choose a project. Review the upstream deployment instructions for the release you plan to pin.

Data dependencies

PostgreSQL for application data. File-dependent features can also need durable storage. See SQL databases for private writer access and connection details.

Deployment order

  1. Create a private PostgreSQL database. Wait for the writer and obtain its host, port and application credentials from Connect.
  2. Set LANGFLOW_DATABASE_URL to the PostgreSQL writer connection. Keep authentication enabled, set an initial admin password securely, and preserve the release’s secret key across deploys.
  3. Use langflowai/langflow on port 7860. Check what the release stores under LANGFLOW_CONFIG_DIR. Provide supported durable storage or use a persistent host if those files must survive task replacement.
  4. Start Langflow after the database is ready. Wait for startup and schema initialization before creating flows. Connect external model and vector services before running any flow that uses them.
Use Runtime variables for runtime credentials. For a Spring Winter web server, set an app-supported health path that returns HTTP 200 without credentials. Check History before running the application check.

Verify the deployment

Sign in, save a small flow, run it with a test input and confirm that the flow survives replacement.