[BUG][Graph] structured_output_model prevents correct output propagation from dependencies #1118
Description
Checks
- I have updated to the lastest minor and patch version of Strands
- I have checked the documentation and this is not expected behavior
- I have searched ./issues and there are no duplicates of my issue
Strands Version
1.14.0
Python Version
3.13
Operating System
macOS 15.6
Installation Method
pip
Steps to Reproduce
Dependencies
python3 -m venv .venv
source .venv/bin/activate
pip install "pydantic>=2.12.0,<3.0.0" "strands-agents[otel]>=1.14.0,<2.0.0" "strands-agents-tools>=0.2.13,<1.0.0"Code snippet
from __future__ import annotations
from typing import Annotated
from pydantic import BaseModel, Field
from strands import Agent
from strands.multiagent import GraphBuilder
from strands.telemetry import StrandsTelemetry
StrandsTelemetry().setup_console_exporter()
class Foo(BaseModel):
foo: Annotated[str, Field(min_length=1, max_length=10)]
bar: Annotated[str, Field(min_length=1, max_length=10)]
first_agent_system_prompt = """
The final output must strictly follow this JSON structure:
``python
class Foo(BaseModel):
foo: Annotated[str, Field(min_length=1, max_length=10)]
bar: Annotated[str, Field(min_length=1, max_length=10)]
``
You must output a valid JSON object conforming exactly to the schema above.
Do not include commentary, explanations, or additional text.
**Example Output:**
``json
{
"foo": "hello",
"bar": "world"
}
``
"""
first_agent = Agent(
name="first_agent",
system_prompt=first_agent_system_prompt,
structured_output_model=Foo,
tools=[],
)
second_agent_system_prompt = """
You must return a valid JSON object conforming exactly to the schema above.
Do not include commentary, explanations, or additional text.
**Example Output:**
``json
{
"foo": "hello",
"bar": "world"
}
``
"""
second_agent = Agent(
name="second_agent",
system_prompt=second_agent_system_prompt,
structured_output_model=Foo,
tools=[],
)
builder = GraphBuilder()
builder.add_node(first_agent, "first_agent")
builder.add_node(second_agent, "second_agent")
builder.add_edge("first_agent", "second_agent")
builder.set_entry_point("first_agent")
graph = builder.build()
def main():
graph("Hello, world!")
if __name__ == "__main__":
main()Expected Behavior
{
...,
"attributes": {
"content": "[{\"text\": \"Original Task: Hello, world!\"}, {\"text\": \"\\nInputs from previous nodes:\"}, {\"text\": \"\\nFrom first_agent:\"}, {\"text\": \" - Agent: {\"foo\":\"...\",\"bar\":\"...\"}\"}]"
}
}Actual Behavior
{
...,
"attributes": {
"content": "[{\"text\": \"Original Task: Hello, world!\"}, {\"text\": \"\\nInputs from previous nodes:\"}, {\"text\": \"\\nFrom first_agent:\"}, {\"text\": \" - Agent: \"}]"
}
}Additional Context
No response
Possible Solution
No response
Related Issues
No response