OpenAI details Codex sandbox design and permission controls
What happened
On November 18, OpenAI published a technical breakdown of the Codex sandbox. It runs commands in the user's existing environment through operating system mechanisms and limits file system and network access according to permission settings. The built-in :workspace profile allows writes to the workspace and broad file reads, protects directories such as .git and .codex, and disables network access by default, but it can still allow reads of sensitive files or changes to local dependencies. The post covers how the sandbox is implemented on macOS, Linux and Windows and shows how to use codex sandbox to restrict permissions for self-built tools and MCP services.
Written by AI from the coverage · updated 2 days ago
Coverage
Follow the reports to see the story from different sides.
- OpenAI NewsPickOpenAI explains how the Codex sandbox limits file and network access for local commands
The Codex sandbox runs commands inside the user's existing environment and uses OS mechanisms to restrict filesystem and network access according to permission settings. The built-in :workspace profile allows workspace writes and broad reads, protects directories like .git and .codex, and disables network by default, but it can still read sensitive files or modify local dependencies. OpenAI details the macOS, Linux and Windows implementations and shows how codex sandbox applies permission limits to homegrown tools and MCP services.
Heat over time
Heat now 3·Comparable peak 7(Nov 19 03:00)·Comparable change over 24 hours -49%
The trend only compares accounts observed without gaps, so its range may be smaller than the current heat. Hover or tap the chart for each hour; the left and right arrow keys step through it.