Curador de Principais Notícias do Hacker News com IA
Um fluxo automatizado, alimentado por IA, para buscar, resumir e apresentar as principais notícias do Hacker News, incluindo detalhes das histórias, URLs e melh...
5 min read
Run Flow Component
The Run Flow component is designed to execute another flow within your AI workflow. This allows you to modularize your processes, reuse existing flows, or orchestrate complex logic by calling sub-flows as part of a larger pipeline.
What Does This Component Do?
This component takes your specified input and passes it, along with any additional variables, to another flow that you choose. It can run synchronously (waiting for a result) or asynchronously (fire-and-forget). This is especially useful for building scalable, maintainable AI systems where you want to break down logic into reusable building blocks.
Key Inputs
| Input Name | Type | Description | Required | Advanced |
|---|---|---|---|---|
| Flow Name | Dynamic Select | Choose the name of the flow you want to run. | Yes | Yes |
| Input Value | String | The main input value to be processed by the target flow (e.g., a message or data payload). | No | No |
| Flow Variables | Nested Dict | Additional variables to pass into the flow; can be used to parameterize execution. | No | No |
| Async Mode | Boolean | If enabled, runs the flow asynchronously and does not return output immediately. | No | No |
| Tool Name | String | (Advanced) A reference name for the tool when using agents. | No | Yes |
| Tool Description | String | (Advanced) Description to help agents understand how to use this tool. | No | Yes |
| Trigger Type | Dynamic Select | (Advanced) Specify a trigger for running the flow. | No | Yes |
| Verbose | Boolean | (Advanced) Enable detailed logging for debugging or monitoring. | No | Yes |
Input Details
- Flow Name: The core configuration; select from existing flows in your workspace.
- Input Value: The actual data or message you want the target flow to process.
- Flow Variables: Key-value pairs for passing extra data/parameters into the flow. If used with a Loop component, these may be overridden.
- Async Mode: If enabled, execution is non-blocking and no outputs are returned immediately (useful for long-running tasks).
- Tool Name & Description: Useful when exposing this flow as a tool to agents (for example, in LLM-powered agents).
- Trigger Type: Allows you to define how the flow should be triggered.
- Verbose: For advanced users who want more insight into the flow execution.
Outputs
The component provides multiple outputs, making it versatile for varied workflow needs:
| Output Name | Type | Description |
|---|---|---|
| Run Outputs | Message | The result returned by the flow (if run synchronously) |
| Flow Execution | FlowExecution | Detailed execution object for advanced tracking |
| Flow as Tool | Tool | Exposes the flow as a callable tool for agents |
- Run Outputs: Contains the output message or result from the target flow.
- Flow Execution: Provides execution metadata, useful for monitoring or chaining flows.
- Flow as Tool: Makes the flow available as a tool, for use by AI agents or other components.
Typical Use Cases
- Modularization: Break complex logic into smaller, manageable flows and invoke them as needed.
- Orchestration: Chain multiple flows together to build end-to-end AI processes.
- Tool Integration: Expose entire flows as reusable tools for agents, improving flexibility and capability.
- Asynchronous Processing: Offload time-consuming tasks to run in the background, improving system responsiveness.
Why Use the Run Flow Component?
- Reusability: Avoid duplicating logic by reusing flows across different workflows.
- Scalability: Manage complex AI systems by organizing logic into callable, self-contained units.
- Agent Compatibility: Seamlessly integrate flows as tools for AI agents to use dynamically.
- Flexibility: Pass custom variables, trigger flows in different ways, and choose between synchronous or asynchronous execution.
Summary Table
| Feature | Benefit |
|---|---|
| Run sub-flows | Modular, maintainable workflows |
| Pass variables | Dynamic, parameterized flow execution |
| Multiple outputs | Integrate results, track execution, or provide agent tools |
| Async mode | Non-blocking, scalable processing |
| Agent/tool support | Integrate with LLM-powered or agentic workflows |
This component is essential for anyone looking to build scalable, maintainable, and modular AI workflows. It enables effective composition of flows and seamless tool-based integrations within AI-driven systems.
