LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|Google's AI gives evasive answers about who won the 2020 US election28/10/26 · Google|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|OpenAI and ServiceNow deepen partnership on enterprise workflows28/10/26 · OpenAI|Failed startups' data finds a second life training AI models28/10/26|Anthropic expands AI for Science program to 20 foundations and research institutions28/10/26 · Anthropic|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|
PluginRisingofficial publisher

Codex CLI (OpenAI) et ses extensions

Plugin : A set of capabilities installed into an AI application or a host · requires codex

OpenAI's terminal coding agent, with sandbox, skills, and integrations.

OpenAI's answer to Claude Code; a serious sandbox on macOS.

Official site GitHub · 130,820 ★
Overall score
7.7 /10
Security: one caveat, see the card
Utility
8.0
Reliability
10.0
Security
One caveat
Ease
5 min · Technical

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Codex CLI addresses the need to run a coding agent directly from the terminal, without going through a web interface or a graphical editor. It inspects files in the local repository, proposes edits, runs commands, and automates repetitive tasks, all within a sandbox that is secure by default.

In practice, you run the codex command inside a project directory, sign in with your ChatGPT account or an API key, then describe the task in plain language. The agent reads the code, generates diffs, submits them for approval, and can chain multiple steps within the same session. Built-in commands such as /review, /permissions, and /model let you adjust behaviour on the fly.

The tool is aimed at developers comfortable with the terminal who want to integrate an AI agent into their scripts, CI pipelines, or code review workflows. It can also be used non-interactively via codex exec, making it suitable for automation.

It is less suited to people who prefer a graphical interface or who are new to command-line tools. Full access to the most capable models and certain advanced features requires a paid ChatGPT subscription or API usage billing.

What it offers

  • ·Inspection and editing of local repository files directly from the terminal, with diff display before applying changes.
  • ·Sandbox enabled by default to restrict agent actions without explicit user approval.
  • ·/review command to analyse uncommitted changes, a commit, or a base branch without modifying the working tree.
  • ·Support for skills and plugins to connect Codex to team tools (GitHub, Gmail, Google Calendar, etc.) without leaving the CLI. (paid)
  • ·Non-interactive mode via codex exec for integrating the agent into scripts and CI pipelines.
  • ·Model and reasoning effort selection via /model, adjustable mid-session.
  • ·Session resumption via codex resume to continue interrupted work.
  • ·Sign-in with a ChatGPT Plus, Pro, Business, Edu, or Enterprise account, or via pay-per-use API key.

In practice

Price
Freemium
Commercial use
Commercial use: yes
Available on
macos, linux, cli
Host
codex
Licence
Apache-2.0
FR interface
No

Security and data

Rating
Security: one caveat, see the card
Your data
Data with the vendor
Maintenance
Active
Last release
23 October 2026

Links and repository checked on 25 October 2026 · added on 1 April 2025

How to install Codex CLI (OpenAI) et ses extensions

1. Open a terminal on macOS or Linux. 2. Run the install command: curl -fsSL https://chatgpt.com/codex/install.sh | sh (or npm install -g @openai/codex / brew install --cask codex). 3. On Windows, use: powershell -ExecutionPolicy ByPass -c "irm https://chatgpt.com/codex/install.ps1 | iex". 4. Navigate to a project directory and run codex. 5. Choose "Sign in with ChatGPT" to use your subscription, or configure an API key if you prefer pay-per-use billing. 6. Security note: review the permissions granted to the agent via /permissions before running it on a production repository.
curl -fsSL https://chatgpt.com/codex/install.sh | sh

In the same category

Whole category Code →