Nodes
Nodes are the functional units that make up a workflow in VIDIZMO Agentic AI. Each node carries out a specific action such as starting a workflow, processing information, applying logic, or interacting with an external service. By arranging and connecting nodes, you define how the workflow behaves from start to finish.
Workflows start when a trigger condition occurs and run sequentially (or in parallel) through the connected nodes to complete a task.
Nodes can perform several functions, such as:
-
Initiate workflow execution
-
Retrieve, modify, or output data
-
Apply decision-making or flow control
-
Connect with AI models or external tools
Key characteristics
| Characteristic | Description |
|---|---|
| Task specific | Each node performs one defined function |
| Configurable | Settings and parameters control node behavior |
| Connectable | Output from one node flows to input of another |
| Reusable | The same node type can be used multiple times in a workflow |
| Visual | Represented as boxes on the Workflow Designer canvas |
Node Types
When browsing the Nodes Library,two types of nodes appear:
-
Trigger Nodes
These start a workflow when a defined event occurs.For example, User Prompt, Triggers when a user submits a question in the chatbot.
-
Action Nodes
These carry out tasks within the workflow, such as perform operations, process data, or interact with services.
-
Transforming or evaluating data
-
Running AI inference
-
Sending results to another system
-
Making logical decisions
-
Executing functions via APIs or tools
Add a Node to a Workflow
Drag a node from the Nodes Library into your canvas. Once added, the Node Configuration panel opens automatically and displays the settings required for that node.
Nodes are organized into categories based on their function, such as:
-
Triggers
-
Outputs
-
Processing
-
Data
-
Embedding
-
Logic
-
AI
-
Integration
-
Tools
-
Control Flow
-
VIDIZMO-specific actions
Configure a Node
Select a node on the canvas to adjust its settings in the Node Configuration panel. Each node includes clearly defined elements:
-
Inputs: Variables that receive data from previous nodes or workflow state. Inputs define what information the node requires to perform its function. You can connect an input to another node's output or reference a workflow variable.
-
Outputs: Variables that store the results produced by the node after execution. Outputs pass data to downstream nodes, enabling the workflow to continue processing. Each output can be connected to one or more inputs on subsequent nodes.
-
Parameters: Options you configure to control node behavior, such as model selection, thresholds, or processing rules.
-
Controls: Node-specific UI elements like toggles, buttons, or dropdown lists that adjust how the node operates.
By default, nodes display only required options that must be configured for the node to function correctly.
Tip: Many fields support Expression mode for dynamic values. Switch a field to Expression mode to reference workflow variables using syntax like
{{state.data.fieldName}}.
Node Configuration Panel
The panel on the right side of the designer displays settings for the selected node. From here you can:
- Rename the node
- Edit node properties
- Update logic and parameters
- Add new conditions or fields (for nodes that support dynamic inputs)
Node actions
Over a node on the canvas view available actions:
- Delete: Remove the node from the workflow
- Duplicate: Create a copy of the node with the same configuration
Node Connectors
Nodes include small circular connection points (●) that represent how data and execution flow through the workflow.
-
Connect an output port from one node to an input port on another.
-
Only ports of the same color should be linked.
-
Connector colors help you understand what type of output is being transferred.
Connector Colors
| Node Type | Connector Color | Purpose | Icon |
|---|---|---|---|
| Tool Nodes | Green | Connects tool-related outputs | ● |
| All Nodes | Orange | Defines the standard flow path | ● |
| Parallel Processing Node | Yellow | To enable the same node to run several instances in parallel. | ● |
Dynamic Connectors
Some nodes have connectors that can be added or removed dynamically based on their configuration. For example, a Switch node starts with default connectors, but you can create additional output paths by selecting Add Item in the Node Configuration panel. Each new connector corresponds to a separate case and routes the workflow to a different branch, allowing the node to handle multiple conditions within a single workflow.
Output Ports
Some nodes have a single output port, which passes data to the next step. For example, a User Prompt node outputs the user's message to the connected node.
Other nodes, such as logic or decision nodes, can have multiple output ports. Each port represents a separate path, allowing the workflow to branch based on conditions or different outcomes.
Node Testing
Node testing validates individual node behavior before running the entire workflow. Testing a node executes it in isolation and displays its input state, output state, execution status, and duration.
Node testing helps you:
- Validate that node logic works as expected
- Debug errors without running the full workflow
- Identify performance bottlenecks
Summary
Nodes define every action a workflow performs. By adding, configuring, and connecting nodes, users can automate processes, apply logic, interact with AI models, and integrate external systems. Understanding nodes and how they connect helps you design efficient, intelligent, and scalable workflows.