Frequently asked questions

When you use an AI assistant, it reads the web pages, emails, and documents you point it at. Prompt injection is when someone hides instructions inside that content — text meant for your AI, not for you. A product review might carry a hidden line telling your shopping agent to buy something, or an email might tell your assistant to forward your data. You never see it. Your AI does, and can act on it.

The extension watches the pages your AI reads in the browser and flags hidden instructions that look like injection attempts before they reach your assistant. It runs at the point where you actually use AI, so protection happens live rather than after the fact. It's the first piece of a broader platform — desktop protection is in development.

Remora inspects page content locally to spot injected instructions. We built it to protect you, not to profile you — the design goal is to look at what your AI is about to read, flag what shouldn't be there, and get out of the way. As we add features we'll keep this page updated with specifics, and our full privacy terms will always be the authoritative source.

Anyone who uses AI tools and doesn't want a web page silently giving their assistant orders. You install it yourself and it works for you — you don't need to be a security team or wait for your company to roll it out.