‘GitLost’ Flaw Lets Attackers Trick GitHub AI Agent Into Leaking Private Repos

A 'GitLost' flaw allows attackers to trick GitHub's AI agent into leaking private repositories by manipulating it with malicious commands via indirect prompt injection attacks. This highlights the vulnerability of AI agents to manipulation. Engineers should be cautious when interacting with AI-powered systems. To mitigate this, use secure and validated inputs when interacting with AI agents.