Miguel Afonso Caetano (@[email protected])
"An indirect prompt injection in an implementation blog can manipulate Antigravity to invoke a malicious browser subagent in order to steal credentials and sensitive code from a userβs IDE. Google Antigravity is susceptible to data exfiltration via indirect prompt injection through the agentic browser subagent. Antigravity is Googleβs new agentic code editor. In this article, we demonstrate how an indirect prompt injection can manipulate Gemini to invoke a malicious browser subagent in order to steal credentials and sensitive code from a userβs IDE. Googleβs approach is to include a disclaimer about the existing risks, which we address later in the article." https://www.promptarmor.com/resources/google-antigravity-exfiltrates-data #CyberSecurity #AI #GenerativeAI #Google #Antigravity #GoogleAntigravity #DataExfiltration #AIBrowser




