Sub Project flow Agent
The Sub Project flow Agent lets you create a new AI agent by reusing an existing flow from your project. Think of it as a shortcut that pulls a pre‑built set of steps (the flow) and turns it into a ready‑to‑use agent that can be added to other workflows.
How it Works
When you add this component to your dashboard, you choose a flow that already exists in your project. The component looks up that flow, extracts the nodes (the individual tasks or steps), and compiles them into a single agent. The agent can then be run in other parts of your automation system. If you tick the Stream option, the agent will send its output back to the dashboard in real time, so you can see the results as they happen.
Inputs
- Agent Node: Select the node that will serve as the starting point for the agent. This tells the component where to begin when it builds the agent.
- Flow Name: Pick the name of the flow you want to reuse. The component will find this flow and use its nodes to build the agent.
- Stream: Check this box if you want the agent to stream its output back to the dashboard. Leave it unchecked for a normal, single‑shot run.
Outputs
- Agent Graph: The component produces a compiled graph that represents the new agent. You can connect this output to other components that need an agent to perform tasks.
Usage Example
- Drag the Sub Project flow Agent onto your canvas.
- In the Agent Node dropdown, choose the node that should start the agent.
- In the Flow Name dropdown, pick the flow you want to reuse.
- If you want to see results as they happen, check Stream.
- Connect the Agent Graph output to a component that will run the agent, such as a “Run Agent” or “Execute Flow” component.
The agent will now run the selected flow whenever the downstream component triggers it.
Related Components
- Run Agent – Executes an agent graph and returns its output.
- Execute Flow – Runs a flow directly without converting it to an agent.
- Agent Builder – Creates a new agent from scratch instead of reusing an existing flow.
Tips and Best Practices
- Keep the flow you reuse simple; complex flows may produce large agent graphs that are harder to manage.
- Use the Stream option only when you need real‑time feedback; otherwise, leave it unchecked to reduce network traffic.
- Double‑check the Agent Node selection to ensure the agent starts at the correct point in the flow.
Security Considerations
- The component only accesses flows that belong to your project, so no external data is exposed.
- If the flow contains sensitive information, make sure the resulting agent graph is only used in secure parts of your workflow.