Does Codex Hooks work in the EU?

Unverified. No explicit evidence found. Last verified 14 Nov 2026.

How much does Codex Hooks cost?

Pricing unknown. The example gist is freely viewable. Use of Codex itself depends on OpenAI's plans, which were not covered in the provided material.

Where was Codex Hooks shown?

In 1 creator video by Matthew Berman. Most recently by Matthew Berman on 14 Jul 2026, at 13:00 in "You aren't using Codex like me...".

Codex Hooks

by OpenAI

GitHub

Pre-tool-use hooks for OpenAI Codex that intercept and block destructive shell commands before they run.

UnverifiedCoding Agents & AutomationApp

Updated 14 Nov 2026EU status checked 14 Nov 2026Last shown 14 Jul 2026

Codex Hooks at a glance

What is Codex Hooks?: Codex Hooks is an app from OpenAI. Pre-tool-use hooks for OpenAI Codex that intercept and block destructive shell commands before they run.
Who makes Codex Hooks?: Codex Hooks is made by OpenAI.
Does Codex Hooks work in the EU?: Unverified. No explicit evidence found. Last verified 14 Nov 2026.
How much does Codex Hooks cost?: Pricing unknown. The example gist is freely viewable. Use of Codex itself depends on OpenAI's plans, which were not covered in the provided material.
Where was Codex Hooks shown?: In 1 creator video by Matthew Berman. Most recently by Matthew Berman on 14 Jul 2026, at 13:00 in “You aren't using Codex like me...”.

Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.

About

Codex Hooks, specifically pre-tool-use hooks, are scripts that run before OpenAI's Codex coding agent executes a tool call. A hook can inspect the pending command and block it. The example discussed by creator Matthew Berman, shared as a public GitHub gist, uses this to stop destructive commands such as deleting the root or home directory.

Berman presented the setup as a response to a report of a GPT 5.6 model deleting a user's files. The hooks act as a deterministic guardrail outside the model, so a dangerous command is refused even if the model proposes it. He also recommends using Codex's 'approve for me' approval mode instead of full access, as a second layer of protection.

The material provided is a user-authored gist and a video description, not official documentation. Details such as the exact configuration format, supported events and Codex version requirements are not confirmed here. Users should check OpenAI's Codex documentation before relying on it.

Key features

Runs before a tool call executes and can block it
Example rules block deleting the root or home directory
Works as a deterministic safety layer outside the model
Intended to be combined with the 'approve for me' approval mode
Shared as a copyable GitHub gist

1 moment

As seen on YouTube

Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.

Matthew Berman14 Jul 2026

Live demo

You aren't using Codex like me...

Pre-tool-use hooks that block destructive commands such as deleting the root or home directory, prompted by a report of GPT 5.6 deleting a user's files. The creator also recommends the 'approve for me' approval mode over full access.

13:00Open on YouTube