Skip to main content
Pre-release

This release is a pre-release and may not be stable for production use.

ARIA (Artificial Intelligence Regulation Interface & Agreements)

License CI Project Status: Alpha GitHub Sponsors

What is ARIA?

ARIA is an open-source framework for defining and enforcing AI participation policies in software projects. It provides a standardized way to specify how AI can interact with your codebase, ensuring clear boundaries and responsibilities between human and AI contributors.

Overview

In an era where AI is increasingly involved in software development, ARIA offers a structured approach to managing AI contributions. Similar to how .gitignore helps manage file tracking, ARIA helps manage AI participation through clear, human-readable policies.

Core Features

  • YAML-based policy definition with AWS-style inheritance
  • Built-in policy templates for common scenarios
  • Policy validation and enforcement tools
  • Integration with popular CI/CD platforms
  • Human-readable policy documentation generation
  • IDE integration for Windsurf, Cursor, and more (coming soon)

Policy Models

ARIA provides several foundational models for AI participation:

GUARDIAN

  • Complete restriction of AI participation
  • Suitable for highly sensitive or regulated projects

OBSERVER

  • AI can only analyze and review code
  • Can suggest improvements without direct modifications
  • Ideal for security-focused projects

ASSISTANT

  • AI can suggest and generate code
  • All contributions require human review and approval
  • Maintains strong human oversight

COLLABORATOR

  • AI can contribute to specific project areas
  • Different rules for different components
  • Granular permission control

PARTNER

  • Maximum AI participation with safety guardrails
  • Human oversight on critical changes
  • Comprehensive testing requirements

Quick Start

# Install ARIA
pip install aria-framework

# Initialize ARIA in your project
ariacli init

# Use a template policy
ariacli template apply assistant

# Validate your policy
ariacli policy validate

# View current permissions
ariacli describe

# Generate IDE rules from policy
ariacli ide rules --ide windsurf

# Generate IDE ignore files
ariacli ide ignore --ide cursor

Policy Example

version: 1.0
model: assistant

defaults:
  allow: []  # Deny-all by default
  require:
    - human_review
    - tests

paths:
  'src/tests/**':
    allow: 
      - generate
      - modify
    require:
      - test_coverage

  'docs/**':
    allow:
      - generate
      - modify
      - suggest

Human Responsibilities

Project maintainers must:

  1. Clearly define AI participation boundaries
  2. Review AI-generated contributions
  3. Ensure policy compliance
  4. Maintain documentation accuracy

Documentation

Getting Started

API Reference

Technical Documentation

CI/CD Integration

Tool Integration

Contributing

We welcome contributions! Please see our Contributing Guide for guidelines.

License

ARIA is licensed under the Apache License 2.0. See our License for details.

Project Status

This project is currently in alpha stage development (v0.1.1-alpha). The core concepts and architecture are established, but many features are still being implemented.

⚠️ Important Notes for Users and Contributors

  • Limited Maintainer Availability: This project is maintained on a part-time basis. Response times to issues and pull requests may be delayed.
  • API Stability: APIs are subject to change without notice during this early stage.
  • Current Focus:
    • Improving IDE integration
    • Enhancing policy enforcement mechanisms
    • Building comprehensive documentation
    • Creating a robust test suite
  • Help Wanted: See our ToDo.md for prioritized tasks where contributions would be most valuable.

For more details on how to contribute, please see our Contributing Guide.

Author

Antenore Gatta

GitHub Topics

When searching for this project, look for these topics:

  • ai-regulation
  • ai-governance
  • ai-policy
  • artificial-intelligence
  • policy-enforcement
  • development-tools

AI Contribution Acknowledgment

In the spirit of transparency and dogfooding our own principles, portions of this project (including code, documentation, and project governance) were developed with the assistance of AI tools. All AI contributions were made under human supervision and review, following the principles outlined in our own ARIA policies.

This acknowledgment serves as a practical example of how AI participation can be transparently disclosed in software projects. For more details, see AI_CONTRIBUTIONS.md.

Sponsorship

ARIA is an open-source project maintained in my free time. If you find this project useful, please consider supporting its development through GitHub Sponsors.

Your sponsorship helps:

  • Maintain and improve ARIA
  • Add new features and integrations
  • Create better documentation
  • Provide support to users

Thank you for your support! ❤️

Metadata

Release files for aria-framework 0.1.1a0

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

Source distribution (sdist)

Source distribution for aria-framework 0.1.1a0
File Size Uploaded
aria_framework-0.1.1a0.tar.gz 77.3 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for aria-framework 0.1.1a0
File Interpreter ABI Platform
aria_framework-0.1.1a0-py3-none-any.whl Python 3 none any Details

Total release size: 110.5 kB

Release files / aria_framework-0.1.1a0.tar.gz

Download URL aria_framework-0.1.1a0.tar.gz
Size 77.3 kB
Tags Source
SHA-256 checksum
How to use checksums
99eb041d21f6ed6311fb47d19d964a7904699ab6b996c36e4c7423194f00b126
BLAKE2b-256 checksum
How to use checksums
31f58dc4b6177ab3d6f222f70d3663975316aaa5efa693892ba5d4801585a6be
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.1.0 CPython/3.13.1

Release files / aria_framework-0.1.1a0-py3-none-any.whl

Download URL aria_framework-0.1.1a0-py3-none-any.whl
Size 33.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
c890ac309288ee307e71c6b1c57122e1f5e64886a15cd89e0536ec742ab503f3
BLAKE2b-256 checksum
How to use checksums
f4796448669242feca95ddd662e06cf4034ea461abd787e51b692802d78b6f5c
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.1.0 CPython/3.13.1

Release history Release notifications | RSS feed

This release

0.1.1a0 This release

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