Skip to main content

a2ui-pydantic

CI codecov Python 3.11+ License: MIT

A Python implementation of the A2UI (Agent to UI) protocol using Pydantic. This library provides type-safe, validated models for building AI agent interfaces that render natively across web, mobile, and desktop platforms.

About A2UI

A2UI is a protocol that enables AI agents to generate rich, interactive user interfaces through declarative component descriptions. Instead of executing arbitrary code, agents send structured JSON messages that clients render using their own native widgets. Learn more at a2ui.org.

Features

  • ✅ Complete A2UI v0.8 Schema Support - Full implementation of the A2UI specification
  • ✅ Type-Safe Models - Built with Pydantic v2 for runtime validation and type checking
  • ✅ 18 Component Types - Text, Image, Button, Form inputs, Layout containers, and more
  • ✅ Message Types - BeginRendering, SurfaceUpdate, DataModelUpdate, DeleteSurface
  • ✅ Data Binding - Support for literal values and data model path references
  • ✅ Comprehensive Validation - Automatic validation of required fields, enums, and constraints
  • ✅ Python 3.11+ - Modern Python with type hints throughout

Installation

uv add a2ui-pydantic

Using pip

pip install a2ui-pydantic

From Source

git clone https://github.com/nurokhq/a2ui-pydantic.git
cd a2ui-pydantic
uv sync --all-groups

Quick Start

Creating a Simple UI Message

from a2ui_pydantic import (
    A2UIMessage,
    BeginRendering,
    SurfaceUpdate,
    SurfaceComponent,
    ComponentWrapper,
    TextComponent,
    ButtonComponent,
    Action,
    Styles,
)

# Create a begin rendering message
message = A2UIMessage(
    beginRendering=BeginRendering(
        surfaceId="main",
        root="header",
        styles=Styles(
            font="Roboto",
            primaryColor="#00BFFF"
        )
    )
)

# Create a surface with components
text_comp = TextComponent(
    text={"literalString": "Welcome to A2UI"},
    usageHint="h1"
)

button_comp = ButtonComponent(
    child="button-text",
    action=Action(name="submit", context=[])
)

update_message = A2UIMessage(
    surfaceUpdate=SurfaceUpdate(
        surfaceId="main",
        components=[
            SurfaceComponent(
                id="header",
                component=ComponentWrapper(Text=text_comp)
            ),
            SurfaceComponent(
                id="button-text",
                component=ComponentWrapper(Text=TextComponent(
                    text={"literalString": "Click Me"}
                ))
            ),
            SurfaceComponent(
                id="submit-btn",
                component=ComponentWrapper(Button=button_comp)
            )
        ]
    )
)

# Serialize to JSON
import json
print(json.dumps(message.model_dump(by_alias=True), indent=2))

Using Data Binding

from a2ui_pydantic import (
    DataModelUpdate,
    DataModelEntry,
    TextComponent,
)

# Update data model
data_update = DataModelUpdate(
    surfaceId="main",
    path="/user",
    contents=[
        DataModelEntry(key="name", valueString="Alice"),
        DataModelEntry(key="age", valueNumber=30),
    ]
)

# Reference data in components
text_with_binding = TextComponent(
    text={"path": "/user/name"}  # Binds to data model
)

Project Structure

a2ui_pydantic/
├── __init__.py          # Public API exports
├── base.py              # Base models (Literal*OrPath, etc.)
├── components.py        # All 18 component types
├── messages.py          # Message types (A2UIMessage, etc.)
├── data_model.py        # Data model structures
├── actions.py           # Action and context models
└── enums.py             # Enumeration types

Development

Setup

# Clone the repository
git clone https://github.com/nurokhq/a2ui-pydantic.git
cd a2ui-pydantic

# Install with dev dependencies
uv sync --all-groups

Running Tests

# Run all tests
uv run pytest tests/

# Run with coverage
uv run pytest tests/ --cov=a2ui_pydantic --cov-report=term

# Run specific test file
uv run pytest tests/test_components.py

Linting and Type Checking

# Run pylint
uv run pylint a2ui_pydantic tests

# Run mypy
uv run mypy a2ui_pydantic tests

# Run bandit (security)
uv run bandit -r a2ui_pydantic

Component Reference

The library supports all A2UI v0.8 components:

Content Components:

  • TextComponent - Text with usage hints (h1-h5, body, caption)
  • ImageComponent - Images with fit modes and usage hints
  • IconComponent - Icons from the A2UI icon set
  • VideoComponent - Video playback
  • AudioPlayerComponent - Audio playback with description

Layout Components:

  • RowComponent - Horizontal layout with distribution/alignment
  • ColumnComponent - Vertical layout with distribution/alignment
  • ListComponent - List with direction (vertical/horizontal)
  • CardComponent - Card container
  • TabsComponent - Tabbed interface
  • DividerComponent - Visual separator
  • ModalComponent - Modal dialog

Form Components:

  • ButtonComponent - Button with actions
  • CheckBoxComponent - Checkbox input
  • TextFieldComponent - Text input (short/long/obscured/number/date)
  • DateTimeInputComponent - Date and time picker
  • MultipleChoiceComponent - Multi-select options
  • SliderComponent - Numeric slider

Message Types

  • BeginRendering - Initialize a new surface with root component and styles
  • SurfaceUpdate - Update or create a surface with components
  • DataModelUpdate - Update the data model for a surface
  • DeleteSurface - Remove a surface

Data Binding

Components support two value types:

  1. Literal Values - Direct values (e.g., {"literalString": "Hello"})
  2. Path References - Data model paths (e.g., {"path": "/user/name"})

This enables dynamic UIs where component values are bound to a data model that can be updated independently.

Resources

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request to the GitHub repository.

Status

This library implements the A2UI v0.8 (Stable) specification. The A2UI protocol is currently in public preview and may evolve. This library will be updated to track specification changes.

Metadata

Release files for a2ui-pydantic 0.1.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for a2ui-pydantic 0.1.1
File Size Uploaded
a2ui_pydantic-0.1.1.tar.gz 24.6 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for a2ui-pydantic 0.1.1
File Interpreter ABI Platform
a2ui_pydantic-0.1.1-py3-none-any.whl Python 3 none any Details

Total release size: 46.1 kB

Release files / a2ui_pydantic-0.1.1.tar.gz

Download URL a2ui_pydantic-0.1.1.tar.gz
Size 24.6 kB
Tags Source
SHA-256 checksum
How to use checksums
7d651cdc41e7068f97b7ee78710bf21a760057397fe488a815e0b0a4a626615d
BLAKE2b-256 checksum
How to use checksums
6a6c4c1e73d68df62c61853e41910ca5a92a433c6f027e5680a5f84672eb35ea
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Dec 26, 2025.

Transparency log

Release files / a2ui_pydantic-0.1.1-py3-none-any.whl

Download URL a2ui_pydantic-0.1.1-py3-none-any.whl
Size 21.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
6a5a421cb65c5abddaa4c706a7a240ebabe44555571ad3ef2f6402bfed65912c
BLAKE2b-256 checksum
How to use checksums
fc9460dda021326ab3966a256a5398d26f39cb1a19467ce907087be97f88244c
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Dec 26, 2025.

Transparency log

Release history Release notifications | RSS feed

This release

0.1.1 This release

2 release files

0.1.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page