> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zerodrift.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Check an Email

> Ask ChatGPT or Codex to check an email with ZeroDrift. See the verdicts and how suggested wording works.

**Install first.** These guides need the ZeroDrift plugin. Install it once from the ChatGPT plugin marketplace and it is available everywhere, in ChatGPT and in Codex. See [ChatGPT and Codex Plugin](https://docs.zerodrift.com/codex-plugin.md) for the steps.

[![](https://docs.zerodrift.com/images/openai-plugin/openai-logo-white.png)
Add to ChatGPT](https://chatgpt.com/plugins/plugin_asdk_app_6af4ea9184388191b9ee1ff948e85d34)

Opens the ZeroDrift listing in the ChatGPT plugin directory. Installing it also enables the plugin in Codex.

Ask the assistant to check an email you wrote. ZeroDrift returns a verdict and each flagged line. This guide covers three cases: a Flagged email, a Passed email, and suggested wording.

ZeroDrift can validate any content that ChatGPT or Codex has access to, and the assistant formats the text it sends. Only that text is checked. ZeroDrift validates. It does not rewrite. Any wording the assistant suggests is the assistant's, and ZeroDrift has not checked it. For how the assistant and ZeroDrift split the work, see [How it works](https://docs.zerodrift.com/codex-plugin.md#how-it-works).

## Before you start

* The ZeroDrift plugin is installed and you are signed in. See [ChatGPT and Codex Plugin](https://docs.zerodrift.com/codex-plugin.md) for setup, and ask for the ZeroDrift status to confirm.
* This guide applies to both distributions. A check you ask for is on request: the result is returned and nothing is stopped.

## Verdicts

| Verdict | Meaning |
| - | - |
| **Passed** | The message goes through. |
| **Warned** | The message goes through. The assistant shows the warnings with the send. |
| **Blocked** | Do-not-send finding in a check before sending. The action is stopped. Hooks version only. |
| **Flagged** | Do-not-send finding in an on-request or skill-directed check. The assistant is told to withhold the send. Nothing technical stops it. |

## Check an email that breaks the rules

Ask the assistant to check the email:

```text theme={null}
Check this email with ZeroDrift: <paste your email>
```

If the email breaks your rules, ZeroDrift returns Flagged with each flagged line, a table of severity, rule and policy, and a short explanation. Severities are **Do not send** and **Caution**.

In ChatGPT, the assistant first asks your permission before it sends text to ZeroDrift. The prompt reads "Allow ChatGPT to use ZeroDrift?", lists the sensitive data it will share, and offers **Deny** or **Allow once**.

![ChatGPT asking Allow ChatGPT to use ZeroDrift, listing a government identifier to be shared, with Deny and Allow once buttons](https://mintcdn.com/zerodrift/WnYfzPOJ4zWh4vgY/images/openai-plugin/chatgpt-permission-prompt.png?fit=max\&auto=format\&n=WnYfzPOJ4zWh4vgY\&q=85\&s=740f0b13c44ad51888c3423c8cac2832)

For a real Flagged table with rule and policy names, see the ChatGPT run in [Check a Social Post](https://docs.zerodrift.com/guides/openai-plugin/validate-social-post.md). The same table format applies to an email.

## Check an email that passes

If the email breaks no rules, ZeroDrift returns "ZeroDrift: Passed — no issues found." In ChatGPT, each of five recent emails returned this verdict, with a ZeroDrift job ID. See [Batch Check and Activity](https://docs.zerodrift.com/guides/openai-plugin/batch-check-and-activity.md) for the screenshot.

## Suggested wording

The assistant can suggest wording after a check. The assistant writes the suggestion, and ZeroDrift only checks it when you ask.

## Limits

* **On request only.** A check you ask for does not stop anything. For checks before a send, see [Pre-Send Email Check](https://docs.zerodrift.com/guides/openai-plugin/pre-send-email.md).
* **ZeroDrift does not rewrite.** Suggested wording is the assistant's.
* **Attachments.** ZeroDrift validates text, not files directly. To check an attachment, ask ChatGPT to extract and format its contents for validation. Only the text ChatGPT sends is checked.
* **Review before you send.** A pass is not a substitute for your own review.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.