ChatGPT connectors, developer mode and MCP: connect ChatGPT and Codex to your website feedback

Connect ChatGPT and Codex to your website feedback over MCP: what ChatGPT connectors and developer mode are called now, and the ChatGPT MCP setup that works.

On November 11, 2026, we asked ChatGPT: “How do I give ChatGPT Codex visual feedback on a web page it built?” The answer named no tool, and its only citations were OpenAI's own help pages. A ChatGPT MCP server is the working answer to that question, and it didn't come up.

People are looking for it. In the US, Google sees about 1,300 searches a month for “codex mcp”, 880 each for “chatgpt mcp” and “chatgpt developer mode”, and 590 for “chatgpt connectors” (search data from November 2026). Those are four names for one job: getting something the model can't see into the conversation. For a web page, that something is what the reviewer saw.

What can't ChatGPT see on a page it built?

ChatGPT and Codex can write a page and change it. What they don't have is the review. The comment a client left on the pricing card on Tuesday, the element it was pinned to, the browser it was seen in, the console error that fired just before: none of that is in the repository or the chat. It sits wherever the feedback was collected.

Pasting screenshots into the chat is the manual way across. It works for one comment. MCP is the wired way, and it works for the fortieth.

What are ChatGPT connectors, developer mode and MCP?

OpenAI has renamed things since these searches became habits. This is what each term means in OpenAI's own docs, checked November 12, 2026.

MCP

OpenAI's MCP page puts it in one line: “Model Context Protocol (MCP) lets ChatGPT and Codex use tools and information from other systems.” An MCP server is the other system's side of that. It lists what the model may do, such as look up a thread or post a reply, and answers when the model asks.

Connectors

What people still search for as connectors, OpenAI's docs now call plugins. The plugin architecture page defines them: “Plugins are the packages people discover, install, share, and publish in ChatGPT and Codex.” A plugin can hold skills (written instructions), an MCP server, or both.

You can watch the rename in one address. OpenAI's admin guide at /enterprise/apps-and-connectors is now titled “Plugin controls”. It says a plugin can include apps, and that “Apps use Model Context Protocol (MCP) to connect to services.” In the API docs, the built-in connectors sit under the heading Legacy connectors, with connector_id deprecated for models released after September 1, 2026. So the thing you install is a plugin, and the part that talks to another service is an MCP server.

Developer mode

Developer mode is about adding a server of your own, one that isn't in the directory. OpenAI's current steps for that don't use the words at all: open ChatGPT Plugins, select the plus button, choose “Add custom MCP server”, enter the server's address, and finish with “Create as a plugin”.

The term survives in workspace settings. The Plugin controls guide lists a member permission, “Create plugins with MCPs”, described as “Use developer mode to create custom app connections.” The same steps page says “Account and workspace policies apply to adding and using custom MCP servers,” and OpenAI's developer pages don't list which plans that covers. In a company workspace, the person to ask is the admin.

If the tool you want is already in the plugin directory, you need none of this.

ChatGPT MCP setup: how do I add Skyelight?

Skyelight is a website feedback tool: reviewers click an element on a live page and leave a comment on it. It's in ChatGPT's plugin directory, so there is no server address to paste and nothing to switch on first.

  1. Open Skyelight in ChatGPT. Use the directory listing, or search for Skyelight in ChatGPT's plugins.
  2. Add it and sign in. ChatGPT opens Skyelight so you can sign in and approve the connection.
  3. Ask about your feedback. For example, “What feedback is still open on my Skyelight projects?”

ChatGPT asks you to confirm before anything that changes a thread, such as a reply or a status change. If you added Skyelight earlier as a custom connector, remove that copy so ChatGPT doesn't see two. You need an owner, admin or collaborator role in the workspace; a reviewer can't use agent tools. The rest is in Connect your agent.

How do I add a Codex MCP server for the same feedback?

ChatGPT on the web and Codex on your machine are set up separately. OpenAI's MCP page says so directly: “ChatGPT web doesn't read local Codex configuration files.” For Codex, Skyelight's installer does it in one command:

npx @skyelight/mcp init --client codex

It shows the change it will make and writes nothing until you confirm. Restart Codex, and the first time it calls Skyelight a browser window opens for you to sign in and approve. No token is written to disk.

To edit the file yourself, add this to ~/.codex/config.toml:

[mcp_servers.skyelight]
url = "https://app.skyelight.ai/mcp"
auth = "oauth"

[mcp_servers.skyelight.oauth]
client_id = "RbF6Ga0l6S49Qsvu"
callback_url = "http://127.0.0.1:54545/callback"

That is the format OpenAI documents: a [mcp_servers.<name>] table with a url, and an oauth table under it. Two details from the same page are worth knowing. “The ChatGPT desktop app, Codex CLI, and IDE extension share this configuration,” so you set it up once. And a .codex/config.toml inside a repository scopes a server to that project, for trusted projects only. In the Codex terminal, /mcp lists the servers that are active.

Working in Claude Code or Cursor instead? The setup for those is in how to send website feedback to a coding agent.

Give ChatGPT the reviewer's view

Reviewers press C and click the element they mean. ChatGPT and Codex read that thread over MCP and reply on it when the work is done.

What should I ask it?

You don't name tools. Say what you want and the agent picks them. In a ChatGPT chat, the useful asks are about reading and sorting. In Codex, with the repository open, they are about fixing.

  • “What's new on the checkout project since I last looked?” New threads, new replies, and what was resolved or deferred. It keeps a bookmark, so the next ask starts where this one ended.
  • “Write a hand-off document for the marketing site project. Quote the threads and cite their ids.” A review you can check, because every statement traces back to a thread.
  • “Work through the Skyelight items assigned to me. For each one, fix it, reply on the thread with what you changed and the PR link, then resolve it.” The Codex one. It reads each thread with its page and element, changes the code, and answers where the reviewer will see it.
  • “Before you change PricingTable.tsx, is there any open feedback on it?” This needs Skyelight Build, which records the file and line behind each element.

More of these, with the tools each one runs, are in Working with an agent.

Should it fix the thread or just look into it?

A teammate decides that on the thread, before the agent reads it. From My inbox, the Hand off menu sets one of three tasks. Investigate asks the agent to read the thread and the page and report what is going on. Plan asks what the change is and what it touches. Fix asks it to make the change and open it for review. The first two end in a reply and no code. See agent hand-off.

One thing Skyelight doesn't do is tell the agent when new work arrives. Start a session by asking what's new.

When is this the wrong setup?

When nobody has reviewed the page. The MCP hands ChatGPT what people wrote on it. It does not make ChatGPT look at the page and form its own opinion, and an empty project gives it nothing to read. And if your feedback already lives in another tracker, look for that tool in the plugin directory first. Moving the comments to suit the agent is the wrong way around.

Frequently asked questions

Do I need ChatGPT developer mode to connect Skyelight?

No. Skyelight is in ChatGPT's plugin directory, so you add it from the Plugins tab and sign in. In OpenAI's docs, developer mode now appears as part of a workspace permission for creating your own MCP connections, which you only need for a server that isn't in the directory.

What are ChatGPT connectors called now?

OpenAI's current docs use plugins and apps where people still search for connectors. A plugin is the package you install in ChatGPT or Codex, and it can include an app that reaches a service over MCP. In OpenAI's API docs, the built-in connectors now sit under the heading Legacy connectors.

How do I add an MCP server to Codex?

Add a [mcp_servers.<name>] table to ~/.codex/config.toml with the server's url. For Skyelight, npx @skyelight/mcp init --client codex writes that entry for you and shows the change first. OpenAI's docs say the ChatGPT desktop app, Codex CLI and IDE extension share the same file.

Can ChatGPT change my feedback threads without asking?

No. ChatGPT asks you to confirm before anything that changes a Skyelight thread, such as a reply or a status change. Anything it posts appears under your name, because the agent signs in as you and has your access.

Is the Skyelight MCP free to use with ChatGPT and Codex?

Yes. Every Skyelight plan includes every MCP tool, and on Free your agent gets 25 agent threads a month. Each conversation counts one (in ChatGPT, one a person a day) and each thread the agent works counts one more. Pro and above have no limit.

ChatGPT can build the page in a minute. What's wrong with it is still something a person has to point at.

Let ChatGPT read what your reviewers saw

No card required.