ORCFLO Documentation

Connections define how data flows through your workflow and which nodes execute in what order. ORCFLO supports two connection types: regular data connections and conditional connections (true/false branches from Decision Point steps).

Creating Connections

Creating a connection is as simple as clicking and dragging between two nodes on the editor canvas.

Locate the output handle

Find the small circle on the right side of the source node.

Click and drag

Click the output handle and drag toward the target node. A curve follows your cursor.

Drop on input handle

Release on the input handle (left side) of the target node to create the connection.

Connection Types

ORCFLO supports two types of connections, both serving different purposes.

Type	Description	Visual
Regular (Data Flow)	The most common type. Data from the source node's output is automatically available to the target node.	Solid line
Conditional (True/False)	From Criteria Check nodes. Execution follows the true branch (green) or false branch (red) based on condition evaluation.	Green or red line

Conditional Connections

Criteria Check nodes have two output handles: one for the true branch and one for the false branch.

Branch	Position	When Executed
True Branch (Green)	Top output handle	When the condition evaluates to true. Connect to nodes that handle the positive/matching case.
False Branch (Red)	Bottom output handle	When the condition evaluates to false. Use for alternative handling, fallbacks, or the "else" case.

Example: Branching Workflow

Input
  |
  v
If/Else (sentiment == "negative")
  |
  +-- True --> Escalate to Support
  |
  +-- False -> Send Auto-Reply

Managing Connections

Once connections are created, you can modify and organize them.

Action	How To
Delete Connection	Click the connection line to delete it.
Reconnect	Drag from an existing connection's end point to move it to a different node.

Connections are automatically routed as bezier curves for clarity
Multiple connections can come from a single output (fan-out)
Multiple connections can go to a single input (fan-in for parallel merging)
Connections cannot create circular loops (DAG constraint)
Use auto-layout to automatically organize connection routing

Customizing Connections

You can customize the appearance of output connections to improve visual organization and make complex workflows easier to understand at a glance.

Changing Output Line Color

Each node allows you to set a custom color for its outgoing connections. This is particularly useful for:

Visually grouping related data flows
Highlighting critical paths through your workflow
Distinguishing between different types of processing branches
Making large workflows easier to navigate

Select the source node

Click on the node whose output connections you want to customize.

Open the configuration panel

The node configuration panel appears on the right side of the editor.

Find the output color setting

Look for the "Output Line Color" option in the node settings.

Choose a color

Select from the color picker or enter a custom hex value. All outgoing connections from this node will use the selected color.

Color Coding Strategy

Consider establishing a consistent color scheme across your workflows. For example, use blue for data transformation paths, green for success flows, and orange for error handling branches.

Data Flow

Understanding how data flows through connections is key to building effective workflows.

When a node completes, its output is stored and accessible to downstream nodes
Data flows automatically through connections - just write plain text in your task instructions
Only directly connected nodes can access the previous step's output
In DAG mode, nodes run in parallel when all dependencies are satisfied

Execution Order

Nodes execute only after all their input connections' source nodes have completed. This ensures data is always available when needed. In Sequential mode, nodes run one at a time regardless of connections.

Troubleshooting

Problem	Solution
Connection won't create	Ensure you're dragging from output (right) to input (left). Check that the connection wouldn't create a cycle.
Data not available in downstream node	Verify the source node is upstream and connected properly. Check that the connection is complete.
Conditional branch not executing	Confirm you connected to the correct handle (true = top, false = bottom). Test the condition logic.
Nodes running in wrong order	Verify all dependency connections exist. Use auto-layout to visualize the DAG structure.

Connecting Nodes