Critical RCE Vulnerability in Cursor, VS Code, and Google Antigravity Exposes 50 Million Developers
A critical one-click remote code execution vulnerability affecting Cursor, VS Code, and Google Antigravity put an estimated 50 million developers at risk of system compromise.
A critical one-click remote code execution (RCE) vulnerability has been discovered across three widely-used code editors: Cursor, Microsoft VS Code, and Google Antigravity. The flaw, identified by AISLE's AI-powered vulnerability analysis, potentially exposed approximately 50 million software developers to severe system compromise through a single click on a malicious link. While patches have been released for all affected platforms, the incident underscores the rapid spread of security vulnerabilities within AI-native developer tooling.
The exploit mechanism was deceptively simple: attackers could embed a malicious link within a Git commit message. When a developer clicked this link within the editor interface, the application would execute arbitrary code with full terminal privileges. Crucially, this occurred without any user confirmation, warning prompt, or visible indication, providing attackers with a completely covert entry point into a developer's machine. This allowed for the silent exfiltration of sensitive credentials, such as API keys for services like OpenAI, Anthropic, and Stripe, direct installation of persistent malware like keyloggers, and unrestricted access to local files.
AISLE's research team initially detected the vulnerability in VS Code during automated scanning in late 2025. Because Cursor is built upon the VS Code codebase, it inherited the same underlying weakness, making these two popular AI-assisted coding environments vulnerable simultaneously. AISLE responsibly disclosed the issue to both Microsoft and Cursor shortly after its AI system flagged the bug.
The vulnerability later resurfaced in early 2026 within Google Antigravity, Google's new AI-assisted coding environment, which also shares roots with the VS Code architecture. AISLE promptly reported this to Google, which issued a remediation within days. Cursor also acted swiftly to patch the flaw, while Microsoft's fix for VS Code was released at a later date. As of the latest updates, current versions of Cursor, VS Code, and Google Antigravity are no longer affected by this specific vulnerability.
This incident highlights a growing trend in the AI coding tool ecosystem. Many AI-native IDEs are forked from common codebases like VS Code. Consequently, a single vulnerability can silently propagate across multiple products, impacting millions of developers before detection. The very speed and convenience that make these tools attractive for AI-assisted development also accelerate the dissemination of security flaws.
AISLE's findings emphasize the critical importance of continuous, AI-driven vulnerability detection systems, which can identify issues that traditional security tools might overlook. Developers using Cursor, VS Code, or Google Antigravity are strongly advised to update to the latest available versions immediately to ensure their systems are protected.
Given the covert nature of this vulnerability, teams should also consider reviewing recent commit histories and rotating any API keys or credentials that may have been exposed through affected editor versions. This proactive approach is essential to mitigate potential long-term impacts from undetected compromises.