Skip to main content
Arcanflows

Workflow Quickstart

Build your first workflow in under 10 minutes with this step-by-step tutorial.

Overview

In this quickstart guide, you'll create a complete workflow that:

  1. Receives a webhook request
  2. Processes data with an AI agent
  3. Sends a notification with the result

Time required: ~10 minutes

Prerequisites

  • Arcanflows account (free tier works)
  • An AI agent created (or use the default assistant)

Step 1: Create a New Workflow

  1. Navigate to Workflows in the sidebar
  2. Click Create Workflow
  3. Enter a name: "My First Workflow"
  4. Click Create

You'll be taken to the visual workflow builder with an empty canvas.

Step 2: Add a Webhook Trigger

Every workflow starts with a trigger. We'll use a webhook trigger.

  1. Click the + button or drag from the node palette
  2. Select TriggersWebhook
  3. Configure the webhook:
Method: POST
Path: /my-first-webhook
Authentication: None (for testing)
  1. Click Save

Your webhook URL will be displayed:

https://api.arcanflows.com/webhooks/{your-workflow-id}/my-first-webhook

Step 3: Add an AI Agent Node

Now let's process the incoming data with AI.

  1. Click the + on the webhook node's output
  2. Select ActionsAI Agent
  3. Configure:
Agent: Select your agent (or "Default Assistant")
Message: Analyze this request and provide a summary: {{trigger.body.message}}
Wait for response: Yes
  1. Click Save

Step 4: Add a Notification

Let's send the AI response via notification.

  1. Click the + on the AI Agent node's output
  2. Select ActionsSend Notification
  3. Configure:
Channel: Email (or Slack if configured)
To: [email protected]
Subject: Workflow Result
Message:
  Original message: {{trigger.body.message}}

  AI Analysis: {{ai_agent.response}}
  1. Click Save

Step 5: Your Complete Workflow

Your workflow should now look like this:

┌──────────────┐     ┌──────────────┐     ┌──────────────┐
│   Webhook    │────▶│   AI Agent   │────▶│    Send      │
│   Trigger    │     │   Process    │     │  Notification│
└──────────────┘     └──────────────┘     └──────────────┘

Step 6: Test Your Workflow

Option A: Use the Test Panel

  1. Click Test in the toolbar
  2. Enter test data:
json
{
  "message": "Hello! Please analyze this test message and tell me what it's about."
}
  1. Click Run Test
  2. Watch the execution flow through each node
  3. Check your email for the notification

Option B: Use cURL

bash
curl -X POST "https://api.arcanflows.com/webhooks/{your-workflow-id}/my-first-webhook" \
  -H "Content-Type: application/json" \
  -d '{
    "message": "Hello! Please analyze this test message and tell me what it is about."
  }'

Step 7: Publish Your Workflow

Once testing is successful:

  1. Click Publish in the toolbar
  2. Confirm the publish action
  3. Your workflow is now live!

What You Built

Congratulations! You've created a workflow that:

  • ✅ Receives HTTP requests via webhook
  • ✅ Processes data with AI
  • ✅ Sends notifications automatically

Viewing Execution History

To see past executions:

  1. Click Executions tab
  2. View list of all runs
  3. Click any execution to see:
    • Input/output for each node
    • Timing information
    • Any errors that occurred

Next Steps

Now that you've built your first workflow, try these enhancements:

Add Conditional Logic

If AI detects urgency → Send to Slack
Otherwise → Send email summary
  1. Add a Condition node after the AI Agent
  2. Set condition: {{ai_agent.response.urgency}} === 'high'
  3. Connect different notification nodes to each branch

Add Error Handling

  1. Click on any node
  2. Go to Error Handling tab
  3. Configure:
    • Retry on failure: 3 times
    • Fallback action: Send error notification

Store Data

Add a database node to save results:

  1. Add Database node after AI Agent
  2. Configure INSERT operation
  3. Map fields from the AI response

Schedule Regular Runs

Change the trigger to run on a schedule:

  1. Replace webhook with Schedule trigger
  2. Set cron: 0 9 * * * (daily at 9 AM)
  3. Fetch data from an API instead of webhook body

Quick Reference

Common Variables

VariableDescription
{{trigger.body}}Webhook request body
{{trigger.headers}}Request headers
{{trigger.query}}Query parameters
{{node_id.output}}Output from a specific node

Keyboard Shortcuts

ShortcutAction
Cmd/Ctrl + SSave workflow
Cmd/Ctrl + EnterRun test
Cmd/Ctrl + ZUndo
DeleteRemove selected node
Space + DragPan canvas
ScrollZoom in/out

Troubleshooting

Webhook not receiving requests

  • Check the webhook URL is correct
  • Verify the HTTP method matches (POST)
  • Ensure workflow is published

AI Agent timing out

  • Check agent configuration
  • Reduce message complexity
  • Increase timeout in node settings

Notification not sending

  • Verify notification channel is configured
  • Check email/Slack credentials
  • Look for errors in execution history

Example Workflows

Customer Support Router

Webhook → AI Classify → Switch →
  ├─ Billing → Create Billing Ticket
  ├─ Technical → Create Tech Ticket
  └─ General → Send FAQ Email

Lead Qualification

Form Submit → AI Qualify → Condition →
  ├─ Score > 80 → Notify Sales + Create CRM Record
  └─ Score < 80 → Add to Nurture Campaign

Daily Report

Schedule (9 AM) → Query Database → AI Summarize → Send Email Report

Ready to learn more? Check out the Builder Interface for a complete guide to all workflow features.