Chainlit AI framework bugs let hackers breach cloud environments

Two high-severity vulnerabilities in Chainlit, a popular open-source framework for building conversational AI applications, allow reading any file on the server and leak sensitive information. […]