Sasha Sobran
162279358c
Co-authored-by: Bo Yang <ybo@google.com> Co-authored-by: Wei Sun (Jack) <weisun@google.com> Co-authored-by: George Weale <gweale@google.com> Co-authored-by: Swapnil Agarwal <swapnilag@google.com> Co-authored-by: Xuan Yang <xygoogle@google.com> Co-authored-by: Shangjie Chen <deanchen@google.com> Co-authored-by: Yifan Wang <wanyif@google.com> Co-authored-by: Kathy Wu <wukathy@google.com>
40 lines
1.2 KiB
Markdown
40 lines
1.2 KiB
Markdown
# ADK Workflow Sequence Sample
|
|
|
|
## Overview
|
|
|
|
This sample demonstrates how to create a simple sequential workflow with **ADK Workflows**.
|
|
|
|
It connects two LLM agents in a chain. The first agent (`generate_fruit_agent`) is instructed to return the name of a random fruit. The output of this agent becomes the input for the second agent (`generate_benefit_agent`), which then tells a health benefit about that specific fruit.
|
|
|
|
In a sequence, the execution flows unconditionally from one node to the next in the order they are defined.
|
|
|
|
## Sample Inputs
|
|
|
|
This sample does not require any input to run.
|
|
|
|
## Graph
|
|
|
|
```mermaid
|
|
graph TD
|
|
START --> generate_fruit_agent
|
|
generate_fruit_agent --> generate_benefit_agent
|
|
```
|
|
|
|
## How To
|
|
|
|
1. Define the agents or functions that will make up the steps in your sequence.
|
|
|
|
```python
|
|
generate_fruit_agent = Agent(...)
|
|
generate_benefit_agent = Agent(...)
|
|
```
|
|
|
|
1. Pass a tuple of three or more elements to `edges` to define an unconditional sequence starting from the first element and passing through each subsequent node in order.
|
|
|
|
```python
|
|
Workflow(
|
|
name="root_agent",
|
|
edges=[("START", generate_fruit_agent, generate_benefit_agent)],
|
|
)
|
|
```
|