Inside the Codex sandbox
- ID
- 47946
- Status
- summarized
- Published
- 18 Nov 2026, 4:00 PM
- Fetched
- 19 Nov 2026, 3:02 AM
- Provider
- OpenAI News
- Category
- ai-labs
- Original URL
- https://openai.com/index/inside-the-codex-sandbox
- Source URL
- https://openai.com/news/rss.xml
Summary
- Score
- 8.0
- Created
- 19 Nov 2026, 3:03 AM
- Tags
- Audience
- developersvibe_codersai_agent_users
What happened
OpenAI's engineering post explains how the Codex sandbox uses permission profiles to set filesystem and network limits. The built-in :workspace profile allows writes inside a Git workspace while keeping .git and .codex read-only, allows broad filesystem reads under existing OS permissions, and disables network access. It notes the tradeoff: sensitive config files such as API keys elsewhere on the machine may remain readable, while dependency downloads require network access; enforcement is described across macOS, Linux, and Windows, and developers can reuse the sandbox in their own tools.
Why it matters
If you run Codex or build an agent harness, do not assume the default sandbox fully contains secrets: a :workspace session can still read sensitive configs outside the repo, and it cannot download dependencies until you grant network. Pick or define a permission profile per task, keep secrets out of readable paths or add read restrictions, and enable network only for install steps that need it.
Discussion angle
What should the safe default be for coding agents: broad read plus workspace write with no network, or narrower reads with an explicit allowlisted package-install step?