Vigil

Detects sensitive data like emails, passwords, and API keys before they are submitted to AI chatbots.

Vigil protects your privacy when using AI chatbots. It runs entirely in your browser and scans your prompts in real-time, detecting sensitive data before it gets submitted — so you never accidentally share private information with AI services. WHAT IT DETECTS: • Email addresses • Passwords and secret assignments • Credit card numbers • Social Security Numbers • API keys • JWT tokens • Database connection strings with credentials • Private key material • Crypto seed phrases • Phone numbers HOW IT WORKS: • Scans prompt text as you type on supported AI sites • Highlights detected sensitive data directly in the prompt preview • Shows an in-page warning alert when sensitive data is found • Tracks daily detection count and calculates a privacy risk score CUSTOMIZABLE: • 16 built-in detection rules — disable any you don't need • Add your own custom regex rules for company-specific patterns • Pause protection with a countdown timer when needed PRIVACY FIRST: • All processing happens locally in your browser • No data is ever sent to external servers • No analytics, tracking, or telemetry • Minimal permissions — only storage and active tab access SUPPORTED SITES: • ChatGPT (chatgpt.com) • Google Gemini (gemini.google.com) Vigil is open source and free to use. Your data stays on your device — always.