Metadata-Version: 2.4
Name: avenqor-ai
Version: 0.1.0
Summary: A simple utility package for text and AI-related helper functions
Author-email: shubham kumar <shubhamk97251@gmail.com>
License: MIT
Project-URL: Homepage, https://github.com/yourusername/avenqor-ai
Project-URL: Repository, https://github.com/yourusername/avenqor-ai
Keywords: ai,nlp,text,utility
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Dynamic: license-file

# avenqor-ai

A simple Python utility package for text processing and AI-related helper functions.

## Installation

```bash
pip install avenqor-ai
```

## Usage

```python
from avenqor_ai import clean_text, count_tokens_approx, chunk_text, similarity_score

# Clean messy text
clean_text("Hello    World!!\n\n")
# -> "Hello World!!"

# Estimate token count
count_tokens_approx("Hello world")
# -> 2

# Split long text into chunks (for LLM context windows)
chunks = chunk_text("your very long document here...", chunk_size=500, overlap=50)

# Compare similarity between two texts
similarity_score("hello world", "hello there")
# -> 0.33
```

## Functions

| Function | Description |
|---|---|
| `clean_text(text)` | Normalizes whitespace and strips text |
| `count_tokens_approx(text)` | Rough token count estimate |
| `chunk_text(text, chunk_size, overlap)` | Splits text into overlapping chunks |
| `similarity_score(text1, text2)` | Jaccard word-overlap similarity (0-1) |

## License

MIT
