ag2-structured-output

name	ag2-structured-output
description	Get a typed Python value back from an AG2 beta `Agent` instead of free text. Pass `response_schema=` (a Pydantic model, dataclass, primitive, union, `ResponseSchema`, or `@response_schema` validator) and read the parsed result via `await reply.content()`. Use when the user wants validated structured output, classification, extraction, or scoring. Covers `ResponseSchema`, `@response_schema`, `PromptedSchema` (for providers without native structured output), per-turn override, validation retries, and primitive embedding.
license	Apache-2.0

Structured output

When to use

• The user wants a Pydantic model, dataclass, dict, primitive, or union back — not a string.
• They're doing classification, extraction, scoring, normalisation, or anything where downstream code parses the reply.
• They want automatic retry on validation failure.

60-second recipe

from pydantic import BaseModel, Field
from typing import Annotated

from autogen.beta import Agent
from autogen.beta.config import OpenAIConfig

class TicketTriage(BaseModel):
    category: Annotated[str, Field(description="e.g. billing, bug, account_access")]
    urgency: Annotated[str, Field(description="low, medium, or high")]
    summary_one_line: Annotated[str, Field(description="Max 120 characters", max_length=120)]

agent = Agent(
    "triage",
    prompt="You triage support messages. Be conservative with urgency.",
    config=OpenAIConfig(model="gpt-4o-mini"),
    response_schema=TicketTriage,
)

reply = await agent.ask("I was charged twice and can't export reports. Quarter close blocked.")
triage = await reply.content()      # → typed TicketTriage
print(triage.category, triage.urgency)

reply.body is still the raw model text; await reply.content() runs validation and returns the parsed value. If validation fails, content() raises (e.g. pydantic.ValidationError).

Schema types you can pass

Type	What you get
Primitive (int, float, bool)	Bare value, framework wraps in {"data": ...} for the API
dataclass	Instance of the dataclass
Pydantic BaseModel	Instance of the model
Union (int | str, (int, str))	One of the alternatives
dict[K, V], TypedDict	Validated dict
ResponseSchema(...)	Same as above, with explicit name / description for the provider
@response_schema callable	Custom validation/parsing logic
PromptedSchema(inner)	Schema injected into the system prompt for providers without native structured output

ResponseSchema — name your payload

Helps the provider treat the structured output as a named contract:

from autogen.beta import Agent, ResponseSchema

schema = ResponseSchema(int | str, name="ByteWidth", description="Number of bits in one byte.")
agent = Agent("assistant", config=config, response_schema=schema)

@response_schema — custom validation

For clamping, regex cleanup, decoding wrapped JSON, or combining fields:

from autogen.beta import Agent, response_schema

@response_schema
def parse_rating(content: str) -> int:
    """Parse a rating and clamp to 1–5."""
    return max(1, min(5, int(content)))

agent = Agent("assistant", config=config, response_schema=parse_rating)

Multi-parameter form synthesises a JSON object schema from the parameter names:

from typing import Annotated
from pydantic import Field
from autogen.beta import response_schema

@response_schema
def extract_listing(
    title: Annotated[str, Field(description="Product name")],
    price_usd: Annotated[float, Field(description="Price in USD", ge=0)],
    in_stock: Annotated[bool, Field(description="True if it ships now")],
) -> dict:
    return {"title": title, "price_usd": price_usd, "in_stock": in_stock}

The function also participates in dependency injection — Context, Variable, Inject, Depends work the same way as in tools (and don't appear in the JSON schema).

Async validators are supported:

import json

@response_schema
async def fetch_and_validate(content: str) -> dict:
    data = json.loads(content)
    data["validated"] = True
    return data

PromptedSchema — for providers without native structured output

Injects the JSON schema into the system prompt instead of using response_format:

from autogen.beta import Agent, PromptedSchema

agent = Agent("assistant", config=config, response_schema=PromptedSchema(int))

Wraps any inner schema (type, ResponseSchema, @response_schema callable). The validation logic stays the same; only the wire format changes.

Custom prompt template:

PromptedSchema(int, prompt_template="Reply with JSON matching this schema:\n{schema}")

Per-turn override

agent = Agent("assistant", config=config)

turn = await agent.ask("How many seconds in a minute?", response_schema=int)
print(await turn.content())   # 60

turn2 = await turn.ask("Say hello.")     # back to default (no schema)

Pass response_schema=None to drop a schema set on the agent for one call.

Retries

When validation fails, automatically re-ask the model:

result = await reply.content(retries=3)   # initial + up to 3 re-asks
result = await reply.content(retries=math.inf)   # interactive only — could loop forever

The validation error is sent back to the model as a follow-up so it can correct itself.

Primitive embedding (embed)

Bare primitives (int, float, bool, list[T], primitive unions) get wrapped in {"data": ...} by default — most structured-output APIs handle objects more reliably than bare values. content() transparently unwraps. Opt out:

ResponseSchema(int, name="RawInt", embed=False)              # model must produce a bare 42
@response_schema(embed=False)
def parse_rating(value: int) -> int: ...

Going deeper

• Working starter: assets/recipe_builder.py (mirrors code_examples/02) — Pydantic model + @tool + response_schema=.
• Full reference: website/docs/beta/structured_output.mdx — covers every schema type, multi-param @response_schema, Field constraints, PromptedSchema, retries, embedding semantics.

Common pitfalls

• Reading reply.body when you wanted typed output — reply.body is the raw text. await reply.content() does the parsing.
• Forgetting await on content() — it's async; you'll get a coroutine, not the value.
• No description in the Pydantic field — the LLM may guess what to put in each field. Add a Field(description=...) for every non-obvious key.
• Provider doesn't support native structured output — wrap with PromptedSchema(...) rather than fighting the API.
• retries=math.inf in production — will loop forever on a model that can't comply. Use a finite count.
• Per-turn override is single-turn — passing response_schema=int to one ask() doesn't change the agent's default. The next turn returns to whatever was set on the constructor.