Guardrails Hub Types

Common errors

• ModuleNotFoundError: No module named 'guardrails_hub_types.HubManifest'

  cause Attempting to import a type directly from the top-level package instead of its specific submodule.
  fix The types are organized into `models` submodules. Correct the import path to `from guardrails_hub_types.models.hub import HubManifest` (or `models.workflow`, etc.).

• pydantic.error_wrappers.ValidationError: 1 validation error for HubManifest\nname\n  field required (type=value_error.missing)

  cause Attempting to instantiate a Pydantic model with missing or incorrectly typed required fields.
  fix Review the schema for the specific type (e.g., `HubManifest`) to ensure all required fields are present and correctly formatted according to their defined types. Refer to the model's source code or generated schema for details.

Warnings

• gotcha This library is version 0.0.x and is subject to rapid change. The API may not be stable and breaking changes are highly likely in future minor versions (e.g., 0.1.0, 0.2.0) before a 1.0.0 release.
  Fix: Always pin to specific versions (`guardrails-hub-types==0.0.4`) in production environments and review release notes carefully for updates.
• gotcha As a types-only library, `guardrails-hub-types` provides Pydantic models for data validation and structure, but does not contain any runtime logic or execution capabilities. It is intended for schema definition and validation.
  Fix: Do not expect this library to provide functional components; it is a dependency for Guardrails Hub's data management, not an executable part of the Guardrails runtime itself.
• breaking The internal structure of the `models` submodules might change, leading to `ModuleNotFoundError` if you import directly from paths like `guardrails_hub_types.hub` instead of `guardrails_hub_types.models.hub`.
  Fix: Always refer to the latest source code or documentation to confirm exact import paths. Stick to the documented `guardrails_hub_types.models.<submodule>.<Type>` pattern.

Install

• pip install guardrails-hub-types Install latest version

Imports

• HubManifest
  wrong:

  from guardrails_hub_types import HubManifest

  correct:

  from guardrails_hub_types.models.hub import HubManifest

  Types are organized into submodules like 'models.hub' or 'models.workflow'.
• WorkflowManifest
  wrong:

  from guardrails_hub_types.workflow import WorkflowManifest

  correct:

  from guardrails_hub_types.models.workflow import WorkflowManifest

  Ensure to import from the correct 'models' submodule path.

Quickstart

This quickstart demonstrates how to import and instantiate a `HubManifest` using `guardrails-hub-types`. Since it uses Pydantic, data validation is automatic, and `ValidationError` is raised for malformed input, as shown with the invalid data example.

from guardrails_hub_types.models.hub import HubManifest
from pydantic import ValidationError

# Example of creating a HubManifest instance
hub_data = {
    "name": "my-hub-rail",
    "version": "0.1.0",
    "description": "A simple rail for demonstration.",
    "authors": ["John Doe <john.doe@example.com>"],
    "tags": ["demo", "llm"],
    "docs_url": "https://example.com/docs",
    "config": {"type": "llm_rail"},
    "inputs": {"prompt": {"type": "string"}},
    "outputs": {"result": {"type": "string"}}
}

try:
    manifest = HubManifest(**hub_data)
    print("HubManifest created successfully!")
    print(f"Manifest Name: {manifest.name}")
    print(f"Manifest Version: {manifest.version}")
except ValidationError as e:
    print(f"Validation Error: {e}")

# Example of invalid data (missing required field 'name')
invalid_hub_data = {
    "version": "0.1.0",
    "description": "Missing name."
}

try:
    invalid_manifest = HubManifest(**invalid_hub_data)
except ValidationError as e:
    print(f"\nFailed to create invalid manifest as expected: {e}")