Jazzmine logo
Plug & PlayStart GuideGitHub
v0.1.0
Python

jazzmine-security

Moderation, sanitization, and toxicity detection primitives for securing AI input and output in backend services.

Overview

Quickstart scope

Start with the minimal end-to-end usage flow, then add sanitizer variants and reference pages as needed for production hardening.

Install

install.sh
python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
pip install jazzmine-security

Optional GPU-enabled PyTorch install:

install_gpu.sh
pip install torch --index-url https://download.pytorch.org/whl/cu121

Key features

  • Input and output moderation with label and confidence scoring.
  • HTML sanitization for model output before client delivery.
  • Toxicity detector with configurable threshold-based blocking.
  • CSV and PDF sanitizer variants for non-HTML content pipelines.

Quick navigation

Usage quickstart

End-to-end moderation, sanitization, and toxicity flow.

Sanitizer variants

CSV and PDF sanitizer usage snippets.

Reference docs

Detailed security internals, moderators, sanitizers, and detector internals.

Reference map

The sections below map the full security API surface if you need internals beyond the quickstart path.

Moderators

3 pages

Sanitizers

1 pages

Toxic Content Detector

1 pages

Detector Internals

2 pages

Errors

1 pages