Graph
Category: Control
Node id: graph
Executes a DAG (Directed Acyclic Graph) of operations defined by nodes and edges
What it does
- DAG (Directed Acyclic Graph) execution of operations
- Flexible node-and-edge graph structure definition
- Topological ordering for dependency-aware execution
- Support for complex branching and merging workflows
Use cases
- Modeling complex clinical workflows with branching logic
- Building dynamic pipelines with conditional paths
- Representing multi-step processes with shared dependencies
Runtime
- Performance tier:
variable
Input ports
| Name | Type | Required | Description |
|---|---|---|---|
input | text | Yes | Initial input to the graph execution |
Output ports
| Name | Type | Required | Description |
|---|---|---|---|
output | text | No | Graph execution result |
node_count | number | No | Number of nodes in the graph |
edge_count | number | No | Number of edges in the graph |
status | text | No | Execution status |
Settings
Basic
| Setting | Name | Type | Required | Default | Allowed values | Description |
|---|---|---|---|---|---|---|
| Graph Definition | graph | json | Yes | Graph structure with nodes and edges arrays |
Errors
| Code | Description |
|---|---|
GRAPH_REQUIRED | graph configuration object is missing |
NODES_REQUIRED | Graph must have at least one node |
INPUT_REQUIRED | No input was provided for graph execution |
INVALID_EDGE | Edge is missing source or target field |
Runtime notes
- Current implementation is a placeholder — returns graph structure info without executing sub-nodes
- Edges are optional; a graph with nodes but no edges still executes
- output is a descriptive string, not structured data — downstream nodes should not parse it
See Build any workflow for how to put nodes together and Ports, types and data flow for how port types connect.
Something unclear? Send documentation feedback by email.