Checks before you send
Read the way a receiving server would.
Three rings beside Send score your draft for spam, phishing and AI writing, each out of 100. They redraw a moment after you stop typing, while there is still time to change a line.
In short
What is a pre-send email check?
OpenEmail scores the mail you receive for spam, phishing and AI writing. A pre-send check runs those same checks on your own subject and body, so you see how the content reads before it leaves.
How it works
Hover a ring for its verdict
Each ring holds a score out of 100. Hover one to read its verdict and what it flagged, then change that part and watch the ring redraw.
Where each verdict starts
Spam reads as unlikely under 35 and likely from 60. Phishing is worth a second look from 30 and a scam from 60, and AI writing is possible from 35 and likely from 60.
In the assistant, and in your code
The assistant shows the same three on a message before you approve sending it, and code or an agent can run the check first through the REST API, the SDK, the command line or the MCP server.
What you get
In the product today
Scored, then dropped
The draft is scored on our server and nothing is kept. No model is called.
Scored on your words
The rings read your subject and body, so each score moves with the words you change.
Enough words to judge
The AI score starts at 40 words, once there is enough of your writing to read.
Good practice
Getting the most out of it
- 01
Fix what was flagged
The hover names what raised a score. Change that part, rather than rewriting the whole message.
- 02
Keep the volume down
A subject in capitals and piles of exclamation marks both raise the spam score, so write the way you would speak.
- 03
Check before code sends
Have your code or agent call POST /emails/check before it sends, the way the composer checks yours.
Questions
Asked often
Keep going
Works well with
Start
Your domain,
your mail.
Point a domain at OpenEmail and read it in a mailbox built around it. The free plan covers one domain.