THREATNOIR
Subscribe
Back to Feed
VulnerabilitiesApr 21, 2026

Google Fixes Critical RCE Flaw in AI-Based Antigravity Tool

Google patches critical RCE vulnerability in AI antigravity filesystem tool via prompt injection.

Read at source

Summary

Google has fixed a critical remote code execution flaw in its agentic AI product designed for filesystem operations. The vulnerability stemmed from inadequate input sanitization in prompt handling, allowing attackers to escape the sandbox and achieve arbitrary code execution through prompt injection attacks.

Entities

Google (vendor)Antigravity (product)Prompt injection (technology)