Redaxa - Privacy check for AI prompts and public GitHub repos

Hi everyone, I'm Aurelio, the maker of Redaxa. The problem is familiar if you do client work or ship code: the stack trace you paste into ChatGPT still has the API key in it, the email thread you want summarized carries the client's name in every quoted header, and the `.env` file you committed three months ago is still sitting in a public repo. Redaxa is a fixed step before the paste. - The Chrome extension checks your message inside ChatGPT, Claude, Gemini, Copilot and Perplexity before it sends. The web check does the same from any browser, and there's a Windows app with signed auto-updates. - Findings cover personal data, credentials, financial information and secrets. Each one is specific and readable, and you get a safer version of the prompt to copy. Nothing is changed for you; you decide what to keep. - The free GitHub repo check is read-only and never executes project code. It puts exposed keys and tokens first, keeps the values hidden until you reveal them, compares your lockfiles against the OSV database, and lists what it did not inspect so a clean result means exactly what it says. Two honest limits: no findings means the enabled checks found nothing, not that the text is safe. And Redaxa never tests a leaked key against its provider, so treat every real match as live until you rotate it. Free tier: 5 prompt checks a day and 3 repo checks a day, no account or card. Pro adds protected terms for the names only you know (clients, codenames), plus local repo checks on Windows. I'd really like to know which thing you most often catch yourself pasting that shouldn't be there. Thanks for looking.