Build with AI.
Go live.

• ↗http

  Call APIs

  HTTP methods, headers, and request bodies.

• ↳request-parser

  Parse requests

  Extract fields from incoming requests and events.

• { }data-json

  Transform JSON

  Select values from payloads and earlier steps.

• $_bash

  Run Bash

  Execute shell commands with environment variables.

• pypython

  Run Python

  Process data and run your own scripts.

• ▤sql

  Query databases

  Read or write to PostgreSQL and MySQL.

• ✉email

  Send email

  SMTP, SendGrid, Postmark, Mailgun, Brevo, Mailjet, and Mailchimp Transactional.

• ◇s3

  Manage objects

  Read, write, inspect, or delete S3-compatible objects.

• ◷wait

  Wait for approval

  Pause for a human decision before continuing.

• ✳agent

  Run AI agents

  Invoke an agent with a pinned configuration and workspace.

workflow_id: webhook-delivery
title: Route a webhook
enabled: true
triggers:
  - type: webhook
    event: order.created

context:
  destination_url: https://api.example.com/orders

steps:
  - step_id: parse
    kind: request-parser
    command:
      extract:
        order: .payload.order

  - step_id: order
    kind: data-json
    dependencies: [parse]
    command:
      from_step: parse
      path: .order

  - step_id: deliver
    kind: http
    dependencies: [order]
    command:
      method: POST
      url: "{{ .context.destination_url }}"
      headers:
        X-API-Key: secret://API_TOKEN
      body: "{{ .steps.order.output }}"

Secrets where you need them. Bash and Python use a step-level env block. HTTP, SQL, email, and S3 use secret://NAME directly in command fields. See the syntax →