fix(agent): Return structured output JSON when AgentResult has no text

[afarntrog]

Description

When AgentResult has no text content but structured_output is present, str() now returns the JSON serialization of the structured output instead of an empty string.

This fixes the following two issues:

1. When running print(agent_result) on an AgentResult while using structured output - the user will see the Agent’s message that invokes the tool with the toolUse. This is demonstrated with the following code:

from pydantic import BaseModel, Field
from strands import Agent


class Foo(BaseModel):
    foo: Annotated[str, Field(min_length=1, max_length=10)]
    bar: Annotated[str, Field(min_length=1, max_length=10)]


agent = Agent(
    system_prompt="You are a helpful assistant that converts data into Foo.",
    structured_output_model=Foo,
)
res = first_agent("Hello")

Output

print(res)
>>> Hello! I need to provide you with a structured output using the Foo tool. Let me create a simple example for you.

New Output:

print(res)
{"foo":"hello","bar":"world"}

2. Output propagation in multi-agent graphs where structured output was being lost:
   In graph.py we are stringifying the result and adding it to the node_input before this change, the result would be empty. Now it'll contain the structured output json result.

        for dep_id, node_result in dependency_results.items():
            node_input.append(ContentBlock(text=f"\nFrom {dep_id}:"))
            # Get all agent results from this node (flattened if nested)
            agent_results = node_result.get_agent_results()
            for result in agent_results:
                agent_name = getattr(result, "agent_name", "Agent")
                result_text = str(result)
                node_input.append(ContentBlock(text=f"  - {agent_name}: {result_text}"))

Changes:

• Modified AgentResult.str() to fall back to structured_output.model_dump_json()
• Added unit test test__str__empty_message_with_structured_output to verify fix
• All existing tests pass, maintaining backward compatibility

Related Issues

#1118

Documentation PR

Type of Change

Bug fix

Testing

How have you tested the change? Verify that the changes do not break functionality or introduce warnings in consuming repositories: agents-docs, agents-tools, agents-cli

• I ran hatch run prepare

Checklist

• I have read the CONTRIBUTING document
• I have added any necessary tests that prove my fix is effective or my feature works
• I have updated the documentation accordingly
• I have added an appropriate example to the documentation to outline the feature, or no new docs are needed
• My changes generate no new warnings
• Any dependent changes have been merged and published

---

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[zastrowm]

For completeness (and my curiosity) can you add the new output of your example code to the PR description