framewise-secureline 0.1.1

A client library for content security detection

Here’s the updated README for your framewise_secureline project with the requested changes:

---

Framewise SecureLine

framewise_secureline is a Python client library for interacting with the SecureLine API. It provides robust content security detection, including identifying sensitive information, prohibited topics, and potential attacks within text.

Features

• Content Analysis: Detect and classify content based on predefined filters such as:

  • Benign: Safe content.
  • Prompt Attacks: Attempts to manipulate the system.
  • Denied Topics: Topics flagged as prohibited.
  • PPI Information: Personal Identifiable Information detection.
  • Word Filters: Inappropriate or flagged language detection.

• Custom Filters: Configure your own denied topics, PPI information types, and word filters.

• Retry Mechanism: Automatically retries failed requests with exponential backoff.

• Timeout Handling: Gracefully handles API timeouts.

• Logging: Comprehensive logging for debugging and monitoring.

---

Installation

You can install framewise_secureline using either Poetry or pip.

Poetry Installation

git clone https://github.com/yourusername/framewise_secureline.git
cd framewise_secureline
poetry install

pip Installation

Install directly from PyPI (once published) or from the GitHub repository:

pip install framewise-secureline

Or, for the latest version from GitHub:

pip install git+https://github.com/yourusername/framewise_secureline.git

---

Usage

1. Initialization

Create a SecureLine client using your SecureLine API key:

from framewise_secureline import SecureLine

client = SecureLine(
    api_key="your-api-key",
    denied_topics="medical diagnoses, competitors",
    ppi_information="SSN, medical records",
    word_filters="profanity"
)

2. Detect Content

Detect security concerns in text:

result = client.detect("This is a test message.")
print("Detection Probabilities:", result.raw_response["probabilities"])

3. Update Filters

Update default filters dynamically:

client.update_filters(
    denied_topics="custom topics",
    ppi_information="custom PPI",
    word_filters="custom filters"
)

---

Examples

Basic Detection

message = "This contains personal information like SSN: 123-45-6789."
result = client.detect(message)

print("Benign Probability:", result.raw_response["probabilities"]["Benign"])
print("Detected Issues:", result.raw_response["probabilities"])

Retry Mechanism

result = client.detect(
    text="This is another test message.",
    retry_count=3,  # Retry up to 3 times on failure
    retry_delay=2.0  # Wait 2 seconds between retries
)

---

Running Tests

Framewise SecureLine uses pytest for testing. Run tests with:

poetry run pytest

Integration Tests

To test with real API requests:

1. Set the SECURELINE_API_KEY environment variable:

   export SECURELINE_API_KEY="your-api-key"
   

2. Run integration tests:

   poetry run pytest -m integration
   

---

Logging

Framewise SecureLine provides built-in logging for debugging and monitoring:

import logging
logging.basicConfig(level=logging.INFO)

client = SecureLine(api_key="your-api-key")
result = client.detect("Log this message.")

---

Project Structure

framewise_secureline/
├── framewise_secureline/
│   ├── __init__.py          # Package initialization
│   ├── client.py            # SecureLine client implementation
│   ├── models.py            # DetectionResult class
│   ├── exceptions.py        # Custom exceptions
├── tests/
│   ├── test_client.py       # Unit and integration tests
├── pyproject.toml           # Poetry configuration
├── README.md                # Project documentation

---

Key Concepts

• Retry Logic: Configurable retry mechanism to handle intermittent network issues.
• Timeout Handling: Avoids hanging on API calls by setting a default timeout.
• Dynamic Filters: Update filters at runtime for customized detection.

---

Contributions

We welcome contributions! Follow these steps to contribute:

1. Fork the repository.
2. Create a new branch for your feature or bugfix.
3. Submit a pull request for review.

---

License

Framewise SecureLine is licensed under the MIT License. See the LICENSE file for details.

---

Support

For questions or issues, please contact [e-mail] or open an issue on the GitHub repository.

---

TODOs

• Add support for batch processing.