AI Anti-Leak

AI Anti-Leak is a privacy-focused web extension designed to protect users from submiting sensitive information to LLMs

AI Anti-Leak is a privacy-focused browser extension designed to protect users from accidentally submitting sensitive or restricted information through their prompts to large language models (LLMs) like ChatGPT, Claude, or Gemini. By allowing users to define a custom list of forbidden words, the extension actively monitors the input fields of LLM interfaces. If a prompt contains any of the specified keywords, the extension prevents it from being sent by disabling the submit button in real-time. This project was created to bring more user control, data awareness, and safety into everyday AI usage.