Feature Request: OpenAI should provide a standalone IDE for Codex and simplify authentication

I am a ChatGPT Plus subscriber because I use Codex professionally to develop a large software project.

My biggest frustration is not the quality of the AI—it is the user experience around authentication and account status.

Today, the development experience is spread across several different systems:

• GitHub

• VS Code

• OpenAI account

• ChatGPT Plus subscription

• API authentication

As a paying customer, I often cannot answer very basic questions:

• Am I actually using my ChatGPT Plus subscription?

• Am I authenticated through ChatGPT or through an API key?

• Is my current limit coming from ChatGPT, the extension, or another service?

• Am I even logged into the correct OpenAI account?

I have spent many hours trying to answer these questions instead of developing software.

From my perspective as a customer, GitHub has become the center of the experience, even though I subscribed to OpenAI—not to GitHub. The current workflow encourages users to go through GitHub and VS Code in ways that make the OpenAI authentication difficult to understand.

Whether or not this is intentional, the result is confusion. Many users may incorrectly conclude that OpenAI is unreliable, when in reality they simply cannot understand which account, subscription or authentication method is currently active.

I believe OpenAI should seriously consider building its own dedicated IDE for Codex.

This IDE should be completely independent, simple and transparent.

For example, it should always display:

• OpenAI account currently connected

• Active subscription (Free, Plus, Pro, Business)

• Authentication method (ChatGPT subscription or API)

• Remaining usage

• Next usage reset

• Active model

• Connection status

This would eliminate an enormous amount of confusion.

Developers subscribe to ChatGPT Plus because they want to build software—not because they want to spend hours diagnosing authentication, GitHub integration, account status, or usage limits.

I believe a dedicated OpenAI IDE would provide a significantly better experience than relying on a complex chain of integrations. It would strengthen the OpenAI ecosystem, reduce support requests, reduce unnecessary AI conversations about authentication, and allow developers to focus on what really matters: writing software.