ContextCypher

AI-Powered Threat Modeling and Security Analysis Platform

Overview

ContextCypher is a privacy-focused threat modeling platform that helps security professionals and developers create architecture diagrams and generate AI-powered security analysis. It is offline-first with local Ollama support, and can also connect to cloud AI providers.

Key Features

Free version

• Full diagram creation and editing
• Node types and security zones
• Chat-based diagram analysis
• Save/load diagrams (JSON)
• Local Ollama offline analysis
• OpenAI/Anthropic/Gemini provider support
• 3D diagram visualization
• Comprehensive GRC module with integrated diagram attack paths

Pro version

• AI diagram generation from text
• Advanced threat analysis with MITRE ATT&CK
• Premium themes and effects
• Automatic data sanitization
• Compliance logging and audit trails
• Full conversation history

Installation

Prerequisites

• Node.js 18.x or later
• npm 10.8.2 or later

Install via npm

# Install globally
npm install -g @threatvectorsecurity/contextcypher

# Or with yarn
yarn global add @threatvectorsecurity/contextcypher

Usage

Start the application

contextcypher

The app will start the backend server, then open your browser automatically unless disabled.

First-time setup

1. Open Settings.
2. Choose AI provider:
   • Ollama (recommended for offline use)
   • OpenAI (API key required)
   • Anthropic Claude (API key required)
   • Google Gemini (API key required)

Install Ollama (offline AI)

# macOS/Linux
curl -fsSL https://ollama.com/install.sh | sh

# Pull a model
ollama pull llama3.2

Windows installer: https://ollama.com/download

CLI options

# Start without opening browser
contextcypher --no-browser

# Custom port (macOS/Linux)
PORT=3003 contextcypher

# Custom port (Windows PowerShell)
$env:PORT='3003'; contextcypher

System Requirements

Minimum

• OS: Windows 10+, macOS 10.15+, Ubuntu 20.04+
• RAM: 4GB (8GB recommended)
• Storage: 1GB free
• Browser: Chrome 95+, Firefox 91+, Safari 15.4+, Edge 95+

Local AI (Ollama)

• Additional RAM: 8-16GB
• Additional storage: 10-20GB for models
• GPU: 8GB+ VRAM recommended

Troubleshooting

Port already in use

# macOS/Linux
lsof -i :3001

# Windows
netstat -ano | findstr :3001

Then stop the process or run on a different port.

Ollama connection failed

# Ensure Ollama is running
ollama serve

# Check API
curl http://localhost:11434/api/tags

Permission denied (macOS/Linux)

mkdir ~/.npm-global
npm config set prefix '~/.npm-global'
echo 'export PATH=~/.npm-global/bin:$PATH' >> ~/.bashrc
source ~/.bashrc

Logs

When running in production mode, logs are written to:

• Windows: %LOCALAPPDATA%\\ContextCypher\\logs\\
• macOS: ~/Library/Application Support/ContextCypher/logs/
• Linux: ~/.local/share/ContextCypher/logs/

Updates

# Update to latest
npm update -g @threatvectorsecurity/contextcypher

# Reinstall
npm uninstall -g @threatvectorsecurity/contextcypher
npm install -g @threatvectorsecurity/contextcypher

Support

• Discord: https://discord.com/invite/Ve7gbf2ytc
• Website: https://threatvectorsecurity.com
• Documentation: https://threatvectorsecurity.com
• Support and issues: https://threatvectorsecurity.com/contact/

License

ContextCypher is proprietary software. By using this software, you agree to the applicable Terms of Service.

Security

For security issues, use the contact form and mark the report as a security issue: https://threatvectorsecurity.com/contact/

Legal Notice

No warranty. This software is provided "AS IS" without warranty of any kind. AI-generated content must be validated by qualified security professionals.

Keywords

threat-modeling, security, ai, diagram, architecture, ContextCypher