Your developers use ChatGPT every day. Make sure they don't accidentally share your secrets.
ChatGPT is the most popular AI tool among developers. They paste code snippets, error logs, database queries, and configuration files to get help debugging. These often contain API keys, connection strings, customer email addresses, and other sensitive data. Even with OpenAI's data retention policies, the data still crosses your network boundary.
The AxSentinel Chrome extension scans every message in the ChatGPT interface before it's sent. If PII or secrets are detected, the message is blocked with a clear warning. Regex scanning works instantly in the browser. For ML-powered detection, the extension connects to a local proxy.
Install from the Chrome Web Store. The extension activates automatically on chatgpt.com, claude.ai, gemini.google.com, and other AI chat sites.
For deeper detection beyond regex patterns, run the scanner binary as a local proxy. The extension auto-connects.
~/.axsentinel/sentinel --proxy --port 18900Open ChatGPT and type a message containing a fake SSN (123-45-6789). The extension should block the message with a warning overlay.
Regex patterns are checked in the browser — no proxy needed for common patterns like SSNs, credit cards, and AWS keys.
Scans the message input field in real time. Also works on Claude.ai, Gemini, DeepSeek, Mistral, and Perplexity.
Clean messages send normally. Only messages with detected PII/secrets show a warning overlay.
Connect the local proxy for ML scanning that catches custom tokens and non-standard secret formats.
Free tier includes regex scanning for unlimited developers. Pro adds ML-powered detection and the compliance dashboard.