Automate backend work without hiding it in a one-off script.

Define webhook, event, schedule, or manual triggers and ordered steps in YAML. Each run keeps its status, logs, inputs, outputs, and errors after the original request ends.

Create a workflow Read the workflow guide

workflows/daily-brief.yaml

workflow_id: daily-market-brief
enabled: true

triggers:
  - type: cron
    cron: "0 13 * * 1-5"

steps:
  - step_id: fetch-prices
    kind: http
    command:
      method: GET
      url: https://api.example.com/quotes

  - step_id: publish-brief
    kind: http
    dependencies: [fetch-prices]
    command:
      method: POST
      url: https://dashboard.example.com/api/market-briefs

Example outcome

Publish a daily market brief on schedule.

A cron trigger fetches prices, selects the required JSON fields, and posts the finished brief to the application, with each step recorded in the run.

Keep automation beside the app

Version triggers, shared context, secret references, and steps with the source they support.

Compose work from visible steps

Call HTTP services, transform data, wait, run code, query data, or introduce an approval.

Understand what happened later

Inspect each step instead of reconstructing a failed script from scattered logs.

Why this matters

Operational scripts become invisible until they fail.

Cron jobs, webhook transforms, approval steps, and maintenance tasks are often scattered across scripts and services with no shared run history.

What changes with Adios

Turn events and schedules into inspectable workflow runs

Define triggers, secret-backed context, and ordered steps in YAML, then deploy the workflow from the workspace or CLI.

How it works

Describe the trigger and steps, then inspect every run.

Step 01
Write a workflow manifest or ask the AI agent to create one.
Step 02
Use triggers and steps for HTTP calls, waits, transforms, and approvals.
Step 03
Deploy and inspect workflow runs from the platform.

A real use case

Publish a daily market brief on schedule.

A cron trigger fetches prices, selects the required JSON fields, and posts the finished brief to the application, with each step recorded in the run.

Included in the workflow

01Webhook, event, and schedule triggers
02HTTP steps with secret references
03Approval and wait steps
04Workflow logs and run state

Start here

Turn the next backend script into a workflow you can inspect.

Keep its trigger, steps, secrets, status, and output together long after the original request has finished.

Create a workflow

Keep building

The next pieces usually connected to this one.

View all features

Managed Databases

Create Postgres, pgvector, Redis, MongoDB, MySQL, or RabbitMQ, declare the dependency in requires, and let Adios inject its connection details into the workload.

Explore →

Secrets & Environment Variables

Commit secret:// references so required credential names remain visible while their values are stored separately, then resolved for the runtime or workflow that needs them.

Explore →

Observability & Logs

Build logs, runtime logs, source artifacts, versions, replica health, routes, and workflow runs stay connected in the same deployment path.

Explore →