As cited
Copy frozen at (site build).
ai security
Researchers broke the browsers built into VS Code, Cursor, Copilot, Codex and Claude Desktop
Researchers at Forever Security identified sandbox escape vulnerabilities in the integrated browsers of five artificial intelligence (AI) coding assistants. Cursor permitted arbitrary code execution, while GitHub Copilot, OpenAI Codex, VS Code, and Anthropic Claude Desktop allowed adversaries to access files on the host system. At least two vulnerabilities remain unpatched.
Why it matters: Developers using these AI assistants face code execution and local file disclosure risks when browsing untrusted web pages or opening files; patching status varies across vendors and immediate vendor guidance is needed.
- Source published
- First seen by Cybersecurity Tracker
Source attribution
Correction
Correction recorded as of .
ai security
Researchers broke the browsers built into VS Code, Cursor, Copilot, Codex and Claude Desktop
Researchers at Forever Security identified sandbox escape vulnerabilities in the integrated browsers of five artificial intelligence (AI) coding assistants. Cursor permitted arbitrary code execution, while GitHub Copilot, OpenAI Codex, VS Code, and Anthropic Claude Desktop allowed adversaries to access files on the host system. At least two vulnerabilities remain unpatched.
Why it matters: Developers using these AI assistants face code execution and local file disclosure risks when browsing untrusted web pages or opening files; patching status varies across vendors and immediate vendor guidance is needed.
- Source published
- First seen by Cybersecurity Tracker