Appearance
Examples
Five worked examples, each a plain-language walkthrough of a complete workflow: what it does, which nodes it uses, and how to wire them together. None of these are one-click imports. For four of the five, building one means opening a new workflow in Aerini and following its steps by hand on the canvas; the fifth ships differently, noted below. The examples folder's own README covers the general mechanics of importing a workflow file, for the cases where you do have one.
This page assumes you've read Nodes Reference and Expressions, and it doesn't repeat their field-by-field detail. If an example uses a scheduled or webhook trigger, Background Runs covers what happens once it's running unattended. If it needs an API key, Credentials covers where that goes.
Getting Started: Fetch and Show
The simplest workflow of the five: a Schedule trigger fires every 10 seconds, an HTTP Request node calls a public API, and two Output nodes display the result. A good first build, just to confirm a fresh install works before trying anything real.
This one ships differently from the rest, as a self-hosted Docker deployment package rather than a Markdown walkthrough. See the Examples README for how to unzip and run it, and to recreate the workflow itself on the canvas.
Scheduled Report
Fetches data from an API every weekday morning, extracts one field from the response, and emails the result. Uses a Schedule trigger (cron), HTTP Request, Transform Data, and Send Email.
Webhook Receiver
Accepts an incoming GitHub webhook, checks whether it's a push to main, and posts to Slack only when it matches. Uses a Webhook trigger, If / Condition, Slack, and Stop for everything else.
Includes a note on why the Webhook node's Secret field alone doesn't verify GitHub's own signature, and what to add if that matters for your setup.
AI Pipeline
Runs once a day: fetches the current top Hacker News story, summarizes it with an AI model, and creates a page in Notion with the result. Uses a Schedule trigger (cron), two chained HTTP Request nodes, AI Prompt, and Notion.
Chatbot with AI Memory and Widget
A persistent chatbot: each message from an embedded chat widget gets a reply from an AI model, and the exchange is saved so the next message continues the same conversation instead of starting over. Uses AI Memory to read history before the model runs and write it back after, with an AI Prompt node in between.
The longest of the five. Covers embedding the widget script on a web page, plus security notes on token exposure and prompt injection worth reading before you expose one publicly.
See also
- Nodes Reference, for every field these examples reference
- Expressions, for the
{{...}}syntax wiring node outputs together - Credentials, for the API keys most of these need
- Background Runs, for what happens once a Schedule or Webhook example is running unattended