mirror of
https://github.com/google/adk-docs.git
synced 2026-09-14 16:16:59 +08:00
Juan Carlos Radillo Diaz
87 lines
4.4 KiB
Markdown
87 lines
4.4 KiB
Markdown
# Loop template workflow agent
|
|
|
|
<div class="language-support-tag">
|
|
<span class="lst-supported">Supported in ADK</span><span class="lst-python">Python v0.1.0</span><span class="lst-typescript">TypeScript v0.2.0</span><span class="lst-go">Go v0.1.0</span><span class="lst-java">Java v0.2.0</span>
|
|
</div>
|
|
|
|
The ***LoopAgent*** class is a [template workflow](/agents/workflow-agents/) agent
|
|
that executes its sub-agents in a loop for a specified number of iterations or
|
|
until a termination condition is met. Use the ***LoopAgent*** when your workflow
|
|
involves repetition or iterative refinement, such as revising code or a
|
|
document. As with other templated workflows, the execution of a ***LoopAgent***
|
|
object is not controlled by an AI model, and is deterministic in how it executes
|
|
its sub-agents. The sub-agents within the defined loop may or may not utilize AI
|
|
models, but the overall execution of those sub-agents is ultimately managed by
|
|
the ***LoopAgent*** object you define.
|
|
|
|
!!! note "Alternative: graph-based workflows"
|
|
|
|
Starting in ADK 2.0 for Python and Go, templated workflows have been superseded
|
|
|
|
by more flexible workflow structures, including
|
|
[graph-based workflows](/graphs/) and
|
|
[dynamic workflows](/graphs/dynamic/).
|
|
|
|
### Example scenario
|
|
|
|
You want to build an agent that can generate images of food, but sometimes when
|
|
you want to generate a specific number of items, such as bananas, the agent
|
|
generates a different number of those items in the image, such as an image of 7
|
|
bananas. You have two tools: `Generate Image`, `Count Food Items`. If your goal
|
|
is to keep generating images until it either correctly generates the specified
|
|
number of items, or after a certain number of iterations, you can build your
|
|
agent using a ***LoopAgent*** workflow.
|
|
|
|
### How it Works
|
|
|
|
When the `LoopAgent`'s `Run Async` method is called, it performs the following actions:
|
|
|
|
1. **Sub-Agent Execution:** It iterates through the Sub Agents list _in order_. For _each_ sub-agent, it calls the agent's `Run Async` method.
|
|
2. **Termination Check:**
|
|
|
|
_Crucially_, the `LoopAgent` itself does _not_ inherently decide when to stop looping. You _must_ implement a termination mechanism to prevent infinite loops. Common strategies include:
|
|
|
|
* **Max Iterations**: Set a maximum number of iterations in the `LoopAgent`. **The loop will terminate after that many iterations**.
|
|
* **Escalation from sub-agent**: Design one or more sub-agents to evaluate a condition (e.g., "Is the document quality good enough?", "Has a consensus been reached?"). If the condition is met, the sub-agent can signal termination (e.g., by raising a custom event, setting a flag in a shared context, or returning a specific value).
|
|
|
|
|
|
|
|
### Full Example: Iterative Document Improvement
|
|
|
|
Imagine a scenario where you want to iteratively improve a document:
|
|
|
|
* **Writer Agent:** An `LlmAgent` that generates or refines a draft on a topic.
|
|
* **Critic Agent:** An `LlmAgent` that critiques the draft, identifying areas for improvement.
|
|
|
|
```py
|
|
LoopAgent(sub_agents=[WriterAgent, CriticAgent], max_iterations=5)
|
|
```
|
|
|
|
In this setup, the `LoopAgent` would manage the iterative process. The `CriticAgent` could be **designed to return a "STOP" signal when the document reaches a satisfactory quality level**, preventing further iterations. Alternatively, the `max iterations` parameter could be used to limit the process to a fixed number of cycles, or external logic could be implemented to make stop decisions. The **loop would run at most five times**, ensuring the iterative refinement doesn't continue indefinitely.
|
|
|
|
???+ "Full Code"
|
|
|
|
=== "Python"
|
|
```py
|
|
--8<-- "examples/python/snippets/agents/workflow-agents/loop_agent_doc_improv_agent.py:init"
|
|
```
|
|
|
|
=== "TypeScript"
|
|
```typescript
|
|
--8<-- "examples/typescript/snippets/agents/workflow-agents/loop_agent_doc_improv_agent.ts:init"
|
|
```
|
|
|
|
=== "Go"
|
|
```go
|
|
--8<-- "examples/go/snippets/agents/workflow-agents/loop/main.go:init"
|
|
```
|
|
|
|
=== "Java"
|
|
```java
|
|
--8<-- "examples/java/snippets/src/main/java/agents/workflow/LoopAgentExample.java:init"
|
|
```
|
|
|
|
!!! tip "Global instructions for all agents"
|
|
|
|
Apply consistent rules or identity across all agents in your workflow using `GlobalInstructionPlugin` registered on your `App` or `Runner` object. Do not use the `global_instruction` parameter on `Agent` which is deprecated since ADK Python v1.16.0.
|