Your First Workflow
Zero credentials needed — this tutorial uses only built-in nodes. About 10 minutes.
Understanding the Interface
Canvas
The main area where you build workflows by connecting nodes.Component Palette
Left sidebar with all available nodes organized by category. A few of the categories you will use most:- AI Models - Chat models (OpenAI, Anthropic, Gemini, OpenRouter, Groq, Cerebras, DeepSeek, Kimi, Mistral, Ollama, LM Studio)
- AI Agents - Agent, specialized agents, and memory nodes
- WhatsApp / Telegram / Social - Messaging automation
- Android - Device control
- Code Executors - Python, Monty, JavaScript, TypeScript
- Webhooks and Utilities - HTTP, webhooks, console, chat trigger
- Schedulers - Timer and cron scheduler
Toolbar
Top bar with workflow controls:- File - New Workflow, Open, Export, Import, Copy as JSON
- Start - Run the workflow (triggers begin listening); becomes Stop while running
- Save - Store your changes
- A per-node Run button lives inside each node’s parameter panel for testing one node
Node Basics
Adding Nodes
- Drag and drop from the component palette
- Right-click on canvas to search nodes
Connecting Nodes
- Drag from an output handle (right side) to an input handle (left side)
- Handles are the small circles on node edges
Node Types
Configuring Nodes
- Click a node to select it
- Click the gear icon or double-click to open parameters
- Set values and click Save
Data Flow
Data flows through connections between nodes. Use template variables to access upstream data:Example
If a Webhook Trigger outputs{ body: "Hello" }:
Building a Simple Workflow
Let’s build a workflow that triggers on a schedule and logs the time.Step 1: Add a Cron Trigger
- Drag Cron Scheduler from Schedulers
- Configure:
Step 2: Add a Python Executor
- Drag Python Executor from Code Executors
- Connect Cron output to Python input
- Add this code (
datetimeis pre-injected —importstatements are blocked in the sandbox):
Step 3: Add a Console node
- Drag Console from Utilities (visible in Dev mode)
- Connect the Python Executor’s output to the Console’s input
- The Console prints each run into the Console tab of the bottom panel — that is where you see your workflow working
Step 4: Start it
- Press Save, then the green Start button in the toolbar
- The workflow runs every 5 minutes; watch the results arrive in the Console tab
- Press Stop when you are done
Keyboard Shortcuts
Tips
Next Steps
AI Agent Workflow
Build an intelligent agent with memory
Node Overview
Learn about all node types