Agent with Structured Outputs

output_schema describes the expected type. If parsing or validation fails, result.content can remain a string. Before accessing schema fields in a run result, use isinstance(result.content, YourSchema), replacing YourSchema with the class you passed as output_schema.

JSON mode requests valid JSON. Native structured output sends a JSON Schema with strict_output=True by default; setting it to False relaxes provider schema enforcement. All three examples attempt local Pydantic parsing.

Code

from typing import List

from agno.agent import Agent, RunOutput  # noqa
from agno.models.openai import OpenAIChat
from pydantic import BaseModel, Field
from rich.pretty import pprint  # noqa

class Rating(BaseModel):
    story: int = Field(..., description="Story rating from 1 to 10.")
    acting: int = Field(..., description="Acting rating from 1 to 10.")


class MovieScript(BaseModel):
    setting: str = Field(
        ..., description="Provide a nice setting for a blockbuster movie."
    )
    ending: str = Field(
        ...,
        description="Ending of the movie. If not available, provide a happy ending.",
    )
    genre: str = Field(
        ...,
        description="Genre of the movie. If not available, select action, thriller or romantic comedy.",
    )
    name: str = Field(..., description="Give a name to this movie")
    characters: List[str] = Field(..., description="Name of characters for this movie.")
    storyline: str = Field(
        ..., description="3 sentence storyline for the movie. Make it exciting!"
    )
    rating: Rating = Field(
        ...,
        description="Your own ratings of the movie.",
    )

# Agent that uses JSON mode
json_mode_agent = Agent(
    model=OpenAIChat(id="gpt-4o"),
    description="You write movie scripts.",
    output_schema=MovieScript,
    use_json_mode=True,
)

# Agent that uses structured outputs with strict_output=True (default)
structured_output_agent = Agent(
    model=OpenAIChat(id="gpt-4o"),
    description="You write movie scripts.",
    output_schema=MovieScript,
)

# Agent with relaxed provider schema enforcement
guided_output_agent = Agent(
    model=OpenAIChat(id="gpt-4o", strict_output=False),
    description="You write movie scripts.",
    output_schema=MovieScript,
)

# Get the response in a variable
# json_mode_response: RunOutput = json_mode_agent.run("New York")
# pprint(json_mode_response.content)
# structured_output_response: RunOutput = structured_output_agent.run("New York")
# pprint(structured_output_response.content)

json_mode_agent.print_response("New York")
structured_output_agent.print_response("New York")
guided_output_agent.print_response("New York")

Usage

Set up your virtual environment

uv venv --python 3.12
source .venv/bin/activate

Set your API key

export OPENAI_API_KEY=xxx

Install dependencies

uv pip install -U openai agno

Run Agent

Save the code above as structured_output.py, then run:

python structured_output.py