Layerwatch

Layerwatch catches API keys, tokens, and secrets before they leak into AI chat tools like ChatGPT, Claude, Gemini, and Copilot. THE PROBLEM Developers paste code into AI chats every day. Buried in that code: AWS keys, database credentials, API tokens, private keys. Once you hit send, that secret has left your machine — and no enterprise AI plan gets it back. In 2025 alone, over 29 million secrets were leaked on GitHub, and leaks tied to AI services grew 81% year over year. Most were accidents by developers moving fast. WHAT LAYERWATCH DOES Layerwatch scans what you type or paste into AI chat interfaces — before it's sent — and flags secrets it detects: - AWS access keys and secret keys - API keys and bearer tokens - Private keys and certificates - Database connection strings - Passwords and credentials in config files and .env content When a secret is detected, Layerwatch warns you immediately so you can redact it before it leaves your browser. PRIVACY FIRST — EVERYTHING STAYS LOCAL All detection happens locally in your browser using pattern matching. Layerwatch sends nothing anywhere: - No page content is collected or transmitted - No detected secrets are stored or logged - No browsing data, no analytics, no tracking - No account required Layerwatch cannot leak your data because it never has your data. The extension is open source — you can read every line at github.com/layerwatch. WHO IT'S FOR Developers, DevOps engineers, and anyone who works with code and uses AI assistants. If you've ever pasted a config file into ChatGPT and felt a moment of doubt — this is for you. WORKS ON ChatGPT, Claude, Gemini, Microsoft Copilot, Perplexity, and other AI chat interfaces. FROM THE LAYERWATCH PROJECT Layerwatch is building runtime visibility and control for AI coding agents — see what your AI tools actually access, and stop what they shouldn't. The browser extension is free forever. Learn more at layerwatch.dev