Output validators

Output format validators constrain structured outputs (for example validating a JSON shape after a node runs). Manage them with client.agent_validation.

Upsert from Pydantic models

Pass Pydantic classes (not instances). The client uses each class __name__ as the validator name and model_json_schema() as the JSON Schema.

from pydantic import BaseModel, Field

class Invoice(BaseModel):
    vendor: str = Field(...)
    total: float = Field(...)

result = await client.agent_validation.upsert_validators([Invoice])

Raw JSON Schema dicts

await client.agent_validation.upsert_validators([
    {"name": "Invoice", "json_schema": {"type": "object", "properties": {}}},
])

List and fetch

validators = await client.agent_validation.list_validators(limit=20)
v = await client.agent_validation.get_validator(validators.validators[0].id)

Attach validators to nodes in your agent graph configuration in the product UI or API so failed validation can trigger retries where your platform supports it.

Introduction

Quickstart

Concepts

Clients

Backend

Agents

Tools

Validation

Tutorials

Reference

Upsert from Pydantic models

Raw JSON Schema dicts

List and fetch

​Upsert from Pydantic models

​Raw JSON Schema dicts

​List and fetch

Upsert from Pydantic models

Raw JSON Schema dicts

List and fetch