Metadata-Version: 2.4
Name: receipt-binding
Version: 0.0.1
Summary: Receipt Binding → This package is now `aiir`. Install: pip install aiir
Project-URL: Homepage, https://github.com/invariant-systems-ai/receipt-action
Author-email: "Invariant Systems, Inc." <hello@invariantsystems.io>
License-Expression: Apache-2.0
Keywords: ai,attestation,audit,compliance,inference,integrity,llm,receipts
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Security :: Cryptography
Classifier: Topic :: Software Development :: Quality Assurance
Requires-Python: >=3.9
Requires-Dist: aiir
Description-Content-Type: text/markdown

# receipt-binding

> **This package has been consolidated into [`aiir`](https://pypi.org/project/aiir/).**

## Install the real package

```bash
pip install aiir
```

## What is AIIR?

**AI Integrity Receipts** — lightweight cryptographic commitment records
generated during AI inference that bind model identity, sampling configuration,
and output tokens into a tamper-evident hash chain.

- Zero external dependencies (Python stdlib only)
- Receipt generation in < 100 µs
- SHA-256 hash chains with O(1) amortized cost
- Works with any model format (GGUF, ONNX, cloud APIs)
- MCP server for Claude / Copilot / ChatGPT integration

## Links

- **PyPI**: [aiir](https://pypi.org/project/aiir/)
- **GitHub**: [invariant-systems-ai/receipt-action](https://github.com/invariant-systems-ai/receipt-action)
- **Paper**: [Attestable AI: Cryptographic Receipt-Binding for Local Generative Model Inference](https://arxiv.org/abs/XXXX.XXXXX)

---

*This package (`receipt-binding`) exists to redirect users to the canonical `aiir` package.
Installing it will automatically install `aiir` as a dependency.*
