Nodes
Learn about the building blocks of Deforge workflows and how to use them effectively.
What Are Nodes?
Nodes are the fundamental building blocks of any Deforge workflow. Each node represents a specific function or operation that can be connected to other nodes to create complex AI agents. Think of nodes as individual LEGO pieces that you can combine to build sophisticated structures.
Anatomy of a Node
Every node in Deforge consists of several key components:
Input Ports
These are connection points on the left side of a node that receive data from other nodes. Different nodes have different input requirements, and some nodes (like input nodes) may not have any input ports at all.
Output Ports
These are connection points on the right side of a node that send data to other nodes. The data type of an output port determines which input ports it can connect to.
Configuration Panel
Each node has a configuration panel that allows you to customize its behavior. The available settings depend on the type of node, but might include things like default values, API endpoints, or processing options.
Node Categories
Deforge organizes nodes into several categories based on their function:
Node Properties
Each node has specific properties that determine its behavior and capabilities. These properties include:
Data Types
Nodes work with specific data types, such as Text, Number, Boolean, JSON, Flow, Date, or specialized types like Rag (for knowledge bases). Understanding data types is crucial for creating valid connections between nodes.
Required vs. Optional Inputs
Some node inputs are required for the node to function, while others are optional. Required inputs are typically marked with an asterisk (*) in the node's configuration panel.
Default Values
Many nodes have default values for their inputs or configuration fields. These values are used when no specific value is provided, making it easier to get started with a node.
Using Nodes Effectively
To make the most of Deforge's node system, keep these best practices in mind:
Start Simple
Begin with a small number of nodes and test your workflow frequently. Add complexity gradually as you confirm each part is working as expected.
Use Clear Naming
Give your nodes descriptive names that indicate their purpose in the workflow. This makes it easier to understand and maintain complex agents.
Organize Your Canvas
Arrange your nodes in a logical flow from left to right. Group related nodes together and use comments to document complex sections of your workflow.
Reuse Common Patterns
Save frequently used node combinations as templates or subflows that you can reuse across different projects.
Next Steps
Now that you understand the basics of nodes, you can
Last updated on