Skip to content

Getting Started with GhostAI ​

Welcome to GhostAI! This guide will help you get up and running with secure AI-powered conversations that automatically protect sensitive data.

What is GhostAI? ​

GhostAI is a secure LLM proxy that sits between your users and AI models (OpenAI, Anthropic, etc.), automatically detecting and protecting sensitive information before it leaves your environment.

How It Works ​

┌─────────────┐     ┌─────────────┐     ┌─────────────┐     ┌─────────────┐
│   User      │────▶│   GhostAI   │────▶│   LLM API   │────▶│  Response   │
│   Input     │     │  Detection  │     │  (OpenAI)   │     │  with PII   │
│  with PII   │     │  & Tokenize │     │             │     │  Restored   │
└─────────────┘     └─────────────┘     └─────────────┘     └─────────────┘
                           │
                           ▼
                    ┌─────────────┐
                    │ Audit Log   │
                    │ (Compliance)│
                    └─────────────┘

Key Benefits ​

  • Data Protection: Sensitive data never leaves your environment
  • Compliance Ready: Built-in support for HIPAA, PCI-DSS, GDPR, and 50+ regulations
  • Reversible: AI responses use original values through session-scoped tokenization
  • Audit Trail: Complete logging for compliance verification
  • File Intelligence: Upload files up to 150 MB and ask questions about any part — hybrid search, cross-file retrieval, GPU-accelerated indexing, and PII protection on every retrieved chunk

Choose Your Path ​

Quick Start (5 minutes) ​

Get GhostAI running locally in 5 minutes. Start Quick Setup →

Full Installation ​

Complete installation with all dependencies and configurations. Full Installation Guide →

First Protected Chat ​

Learn how PII detection works with a hands-on tutorial. Try Your First Chat →

System Requirements ​

Minimum Requirements ​

ComponentRequirement
Python3.11 or higher
Node.js18.0 or higher
Memory4GB RAM minimum
Disk2GB free space

Supported Platforms ​

  • macOS: 12.0 (Monterey) or later
  • Linux: Ubuntu 20.04+, Debian 11+, CentOS 8+
  • Windows: Windows 10/11 with WSL2
  • Docker: Any platform with Docker 20.10+

API Keys Required ​

You'll need at least one LLM provider API key:

Next Steps ​

  1. Quick Start - Get running in 5 minutes
  2. Model Comparison - Compare models side-by-side with Consensus, Mega Synthesis, and Combined Answer
  3. Configure Your Vertical - Set up industry-specific detection
  4. Explore Features - Learn about all capabilities
  5. File Intelligence - Hybrid search, cross-file retrieval, GPU embedding
  6. API Integration - Build custom integrations

SOC 2 Type II Certified | BAA Available | HIPAA-Ready