Spam scoring API
Riffraff scores each form submission against a stack of plain rules, adds them up, and tells you which rule cost what. No model, no black box. When it gets one wrong you can see exactly why, and say so.
A real submission, scored
Re: your website
Total
270
170 over the line
That is the whole response. Four rules fired, and each one says what it cost.
What you get
01
POST JSON to /api/evaluate. you get a verdict and a score back. that’s the whole API.
02
every flag ships with a plain-english explanation. no opaque ML. reply to legitimate messages with confidence.
03
install the addon, paste a token, done. submissions get scored automatically and spam is stored with its reasons.
How it works
send the contents of the form. we send back a verdict. what you do with it — deliver, drop, queue for review — is up to you.
Request
01POST /api/evaluate
Authorization: Bearer <your-token>
{
"email": "[email protected]",
"subject": "quick question",
"content": "Hi — do you offer…"
}Response
spam: 270 / 100{
"is_spam": true,
"score": 270,
"threshold": 100,
"reasons": [
"Spam phrases detected: \"business proposal\"",
"Non-professional email domain",
"Bot-generated name pattern"
]
}Pricing
every plan includes the full API, plain-english reasons, and the statamic addon. pick a monthly allowance that matches your traffic.
free
1,000 evaluations / month
£19.00/mo
25,000 evaluations / month
No card to sign up
£99.00/mo
250,000 evaluations / month
No card to sign up
Need more than the pro allowance? Get in touch and we’ll size something up.
one API call. no configuration. free tier, no card.