CLI
epiphany run plans a pipeline from English. Configure by typing in the terminal, then accept and deploy to Epiphany Cloud.
Install
One line. Detects macOS or Linux (Intel or Apple Silicon / ARM) and installs the latest release into /usr/local/bin or ~/.local/bin.
curl -fsSL https://epfny.io/install.sh | sh
Log in
Authenticate with your Epiphany account. Your token is stored automatically in ~/.config/epiphany/config.json.
epiphany login
Plan from English
Same as the hero. Prints workflow and pipeline ids, plan steps, and ml_goal if you mentioned a modeling target.
epiphany run "Nightly S3 orders into Snowflake, train a churn baseline"
Configure in the terminal
No JSON file required. The CLI loads the workflow form schema and prompts for source, destination, schedule, and credentials. Pick a number for dropdowns. Passwords are hidden. Use a saved connection to skip secrets.
epiphany workflows configure 12 # or continue from run: epiphany run "S3 orders → Snowflake" --interactive --accept --deploy
Configure from a file (optional)
Same payload the web form sends as values.
epiphany workflows configure 12 --file examples/configure.json
Accept and deploy
Accept generates ingest + transform. Deploy sends the execution to Epiphany Cloud (Azure Container Apps).
epiphany workflows accept 12 epiphany workflows deploy 12 --cron "0 2 * * *" --cpu 1.0 --memory 2Gi epiphany workflows status <deployment-id>
Labs and monitor
Open Data Lab or ML Lab on a pipeline. List promoted model health.
epiphany labs data --pipeline 3 epiphany labs ml --pipeline 3 epiphany labs monitor