- Introduction
- Getting started
- Process modeling with BPMN
- Process modeling with Case Management
- Designing a persistent case entity schema
- Defining case keys (system vs. external)
- Establishing task I/O and write-back contracts
- Exit rules and early stage termination
- Modeling primary and secondary stages
- Triggering a case from Data Fabric
- Implementing stage-level personas and permissions
- Setting SLAs and automated escalation rules
- Configuring a rework loop (re-entry)
- Managing live case instances: pause, migrate, and retry
- Maestro case management component dictionary
- Process modeling with Flow
- Getting started
- Core concepts
- Node reference
- Built-in nodes
- Script
- HTTP Request
- Data Transform
- Decision
- Switch
- Loop
- Delay
- Merge
- End
- Terminate
- Mock
- Manual Trigger
- Scheduled Trigger
- Agent
- Business Process Model and Notation (BPMN) process
- RPA Workflow
- API Workflow
- Human Task
- Agent tools
- Connector nodes
- Build guides
- Best practices
- Reference
- Process implementation
- Debugging
- Simulating
- Publishing and upgrading agentic processes
- Common implementation scenarios
- Extracting and validating documents
- Process operations
- Process monitoring
- Process optimization
- Reference information
Maestro user guide
Invokes a Business Process Model and Notation (BPMN) process that is already published in your organization — an agent or agentic automation built elsewhere — and returns its result. Use this node to reuse a published process as a step in your flow, rather than authoring an agent inline with the Agent node.
Before you start: publish a process
The BPMN process node has no process of its own. It surfaces the BPMN processes published in your organization, so one must be published before it appears under the node.
- The palette lists published processes under BPMN process, each as its own node.
- If nothing is published, the category appears but is empty.
How it works
Published processes are loaded dynamically from your organization, the same way connector nodes and Extract models are. Adding one to the canvas creates a node bound to that specific process. When it runs, Flow invokes the process, waits for it to complete, and returns its result before continuing.
Configuration
| Field | Required | Description |
|---|---|---|
| Process | Yes | The published BPMN process to invoke. Chosen when you add the node from the palette. |
| Input | No | Values mapped to the process's input schema, bound to workflow data with expressions such as $vars.start.output.request. |
Output variables
| Variable | Description |
|---|---|
output | The process's result, structured according to its output schema. Access fields with $vars.<node>.output.<field>. |
error | Populated if the node fails. See Error handling. |
Agent vs BPMN process
| Use… | When you want to… |
|---|---|
| Agent | Author the agent inline on the canvas — set its model, prompts, tools, and context here. |
| BPMN process | Reuse an agent or agentic automation that is already published in your organization. |
Notes
- To test without invoking the real process, use Mock output on the node and supply a sample result.
- The BPMN process node supports an error handle — connect it to route failures instead of stopping the process.
- The list of processes reflects what is published in your organization at design time. Publish or republish a process, then reopen the palette to see it.