Wei (Jack) Sun
1c64a41b53
Introduces two major capabilities: - Workflow runtime: graph-based execution engine for composing deterministic execution flows for agentic apps, with support for routing, fan-out/fan-in, loops, retry, state management, dynamic nodes, human-in-the-loop, and nested workflows - Task API: structured agent-to-agent delegation with multi-turn task mode, single-turn controlled output, mixed delegation patterns, human-in-the-loop, and task agents as workflow nodes Co-Authored-By: Bo Yang <ybo@google.com> Co-Authored-By: George Weale <gweale@google.com> Co-Authored-By: Sean Zhou <seanzhougoogle@google.com> Co-Authored-By: Shangjie Chen <deanchen@google.com> Co-Authored-By: Swapnil Agarwal <swapnilag@google.com> Co-Authored-By: Wei Sun <weisun@google.com> Co-Authored-By: Xuan Yang <xygoogle@google.com> Co-Authored-By: Yifan Wang <wanyif@google.com> Change-Id: I35932c50cfe29ff68559e3781713dbb5eb7b3382
ADK Workflow Routing Sample
Overview
This sample demonstrates how to use routing in ADK Workflows.
It takes user input and uses an LLM node to categorize it as a question, a statement, or other. Based on the classification, it appropriately routes the execution to a specialized agent or function to handle that specific type of input.
In ADK Workflows, routing allows conditionally executing different execution paths based on the output of a previous node.
Sample Inputs
What is the capital of France?The weather is very nice today.Translate bonjour to english
Graph
[ START ]
|
v
[ process_input ]
|
v
[ classify_input ]
|
v
[ route_on_category ]
/ | \
"question" "statement" "other"
/ | \
v v v
[answer_question] | [handle_other]
|
[comment_on_statement]
How To
-
A node (agent or function) yields an
Eventwith a specific route name:yield Event(route="your_route_name") -
In the
Workflowedges definition, conditional edges are constructed using a routing map dict as the second element of the edge tuple:(source_node, {"your_route_name": target_node})