Metadata-Version: 2.1
Name: reason-py
Version: 0.1.6
Summary: Official Python client for reason:// — the reasoning infrastructure for the agentic internet
Author-email: Jacob Westerbeck <jacob@pcfic.com>
License: CC BY 4.0
Project-URL: Homepage, https://github.com/Astrognosy-Ai/reason
Project-URL: Documentation, https://github.com/Astrognosy-Ai/reason/blob/main/sdk/reason_py/README.md
Project-URL: Bug Tracker, https://github.com/Astrognosy-Ai/reason/issues
Keywords: reasoning,agents,protocol,warf,ai
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: Free To Use But Restricted
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Description-Content-Type: text/markdown

# reason-py

Official Python client for [reason://](https://github.com/astrognosy/reason) — the reasoning infrastructure for the agentic internet.

```bash
pip install reason-py
```

## Quickstart

```python
from reason_py import ReasonClient

client = ReasonClient()

# Resolve a validated reasoning artifact
artifact = client.resolve("reason://finance/fraud/synthetic-identity-temporal-motif")

# Apply it locally — nothing leaves your environment
for transaction in my_transactions:
    similarity = client.compare(transaction.embedding, artifact)
    if similarity > artifact.thresholds.high_confidence:
        flag(transaction)
```

## Point at your own node

```python
client = ReasonClient(endpoint="http://localhost:8080")
```

## What transfers

The artifact contains only the non-invertible structural centroid of a validated pattern. No raw data. No model weights. Reconstruction rate empirically measured at r = 0.0149 (< 1.5% of original signal recoverable).

## License

CC BY 4.0 — Jacob Westerbeck / [Astrognosy AI](https://astrognosy.com)
