Metadata-Version: 2.4
Name: ai_language_filter
Version: 0.1.2
Summary: Language-data authenticity enforcement tool
Author-email: "Nanke Zhou (南柯舟)" <dong@nankezhou.net>
License-Expression: LicenseRef-NKZ
Project-URL: Homepage, https://nankezhou.net
Project-URL: Repository, https://example.com/your/repo
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Environment :: Console
Classifier: Topic :: Scientific/Engineering :: Information Analysis
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE-NKZ.txt
Requires-Dist: PyYAML<7,>=6
Requires-Dist: openpyxl<4,>=3.1
Requires-Dist: regex<2026,>=2024.5
Dynamic: license-file

# ai-language-filter

A Python package to filter prohibited behaviors in AI language models.

Strict filter to forbid synthetic/placeholder/inferred/virtual values, enforce >=2 authoritative-source crosschecks, and produce full audit trails.

## Quickstart
```bash
pip install ai-language-filter
ai-language-filter --version
ai-language-filter --mount ./_alf
ai-language-filter --fail-on-violation "this contains simulated placeholder"


# AI Language Filter

A Python package to filter prohibited behaviors in AI language models.

## Installation
pip install ai-language-filter

## Usage
from ai_language_filter.core import FilterPipeline

pipeline = FilterPipeline(config_path="config.toml")
result = pipeline.process("Synthetic data here {{placeholder}}")
print(result)  # Blocked or redacted output

## Extension
See examples/ for custom plugins.

