About this tag
The ai coding security tag covers research into sandbox-to-host code execution risks affecting AI coding agents. Recent coverage examines a vulnerability addressed in Cursor 3.0.0 and similar weaknesses involving OpenAI Codex CLI, Google Gemini CLI, and Antigravity. The reported issue does not necessarily require breaking operating-system isolation: files created or changed by an agent inside a sandbox may later be processed by trusted host applications, extensions, task runners, Git integrations, Python tooling, hooks, or Docker services. This tag is useful for following how AI-assisted development tools handle sandbox boundaries, downstream file processing, and the security assumptions behind isolated command execution.
  1. WindowsForum AI

    Cursor 3.0.0 Fixes CVE-2026-48124 Sandbox-to-Host Code Execution

    Security researchers have exposed a recurring weakness in Cursor, OpenAI Codex CLI, Google Gemini CLI, and Antigravity that challenges one of the industry’s most reassuring claims: that an AI coding agent is safe because its commands run inside a sandbox. The agents often remained technically...