Preview. These maps are our reading of public sources and haven't been checked with the projects' maintainers yet.
OpenAI Codex sandbox
The OS-enforced sandbox plus approval policy in Codex CLI, IDE extension and the ChatGPT desktop app: Seatbelt on macOS, bubblewrap on Linux/WSL2, native sandbox on Windows; workspace-only writes and no network by default.
openai/codex rust-v0.164.0 released 2026-10-23; repo pushed 2026-10-23. (checked 2026-10-25)
In Agent guardrails
Where it stands apart
- Do you need an account to use it? account required (24% of 21 others)
- Which shared standard or format does it mainly build on? OS sandbox primitives (Seatbelt, bubblewrap, Linux namespaces, seccomp) (14% of 21 others)
- Which agents does it work with? one specific agent (19% of 21 others)
- Does it control the agent's outbound network access? yes, deny by default or allowlist by default (6% of 16 others)
What it shares with most of the community
- Is it actively maintained? active (81% of 21 others)
- How is the main thing licensed? permissive open source (MIT/Apache/BSD) (67% of 21 others)
- How is it mainly sustained? company with a paid product, tier, or ads (50% of 20 others)
- Can it run entirely on your own device or infrastructure, without the maker's servers? yes (76% of 21 others)
- Who is it for? developers (90% of 21 others)
- What mainly decides whether an action is allowed? rules, policy, or boundaries written in advance (57% of 21 others)
- What is the main thing it protects? the developer's machine and files (52% of 21 others)
- Is enforcement kept out of the agent's reach? yes, enforced in code or infrastructure the agent cannot change (95% of 21 others)
Not documented
The public sources we read don't say:
- If the guard itself breaks, is missing, or cannot decide, what happens by default?
Nearest neighbors
- Claude Code sandboxing 88% of 16 shared answers match
- Drop 75% of 16 shared answers match
- E2B 69% of 16 shared answers match
Every answer
| Question | Answer | Source |
|---|---|---|
| Is it actively maintained? | active | source 2026-10-25 |
| How is the main thing licensed? | permissive open source (MIT/Apache/BSD) Codex CLI including the sandbox code is Apache-2.0. | source 2026-10-25 |
| How is it mainly sustained? | company with a paid product, tier, or ads Part of OpenAI's paid ChatGPT/Codex plans. | source 2026-10-25 |
| Can it run entirely on your own device or infrastructure, without the maker's servers? | yes CLI and sandbox are local; Codex can use local models (--oss), though the default is OpenAI's hosted models. | source 2026-10-25 |
| Do you need an account to use it? | stands apart account required Codex sign-in is with ChatGPT account or OpenAI API key; sandbox is part of Codex. | source 2026-10-26 |
| Who is it for? | developers | source 2026-10-25 |
| Which shared standard or format does it mainly build on? | stands apart OS sandbox primitives (Seatbelt, bubblewrap, Linux namespaces, seccomp) Seatbelt, bubblewrap, native Windows sandbox. | source 2026-10-26 |
| Where does it mainly step in? | at the OS or VM boundary (sandbox, container, syscall supervisor) | source 2026-10-25 |
| What mainly decides whether an action is allowed? | rules, policy, or boundaries written in advance Sandbox boundary set in advance; an approval policy asks a person to cross it, or routes to an auto-review reviewer agent. | source 2026-10-25 |
| If the guard itself breaks, is missing, or cannot decide, what happens by default? | Not documented Docs say Codex 'surfaces a startup warning when bwrap is missing'; did not confirm whether commands then run unsandboxed. | source 2026-10-25 |
| Which agents does it work with? | stands apart one specific agent | source 2026-10-25 |
| In the default setup, where does the agent's work actually execute? | in a local VM, container, or OS sandbox | source 2026-10-25 |
| What is the main thing it protects? | the developer's machine and files | source 2026-10-25 |
| Does it keep a record of agent actions? | no, or not a focus | source 2026-10-25 |
| Does it control the agent's outbound network access? | stands apart yes, deny by default or allowlist by default 'By default, the agent runs with network access turned off.' | source 2026-10-25 |
| What is it mainly for? | letting agents run with fewer approvals, safely 'The sandbox reduces approval fatigue.' | source 2026-10-25 |
| Is enforcement kept out of the agent's reach? | yes, enforced in code or infrastructure the agent cannot change | source 2026-10-25 |
Sources
This page was derived from:
- GitHub · github.com/openai/codex · read 2026-10-25
- Web page · developers.openai.com/codex/agent-approvals-security.md · read 2026-10-25
- Web page · learn.chatgpt.com/docs/sandboxing · read 2026-10-25
- Web page · learn.chatgpt.com/docs/sandboxing.md · read 2026-10-25
Are you a maintainer? Claiming and correcting projects is coming soon.