Evally
Trust & safety

Spam detector

Separate real messages from promotion, bait and bots, and only delete what's clear.

Scores promotion, link-pushing, copy-paste genericness and bot patterns against signs of a genuine contribution. Delete only the spam it's sure about and send the grey zone to a person.

Try it live

Real calls to Jev. Pick an example or paste your own.

Examples
91
No verdict yet

Pick an example or paste your own input, then run it.

Under the hood

Every question and rule, in the open. Fork it to change any of them.

What it asks

Reads message, then asks 5 questions in one call.

  1. Q1PromotionYes / no

    Is the message unsolicited promotion for a product, service, channel or crypto project?

  2. Q2Pushes a linkYes / no

    Does the message push the reader to click a link, join a group, or move to another app?

  3. Q3GenericYes / no

    Is the message generic enough to have been sent to thousands of people unchanged?

  4. Q4Bot patternsYes / no

    Does the message look automated (random characters, repeated phrases, keyword stuffing)?

  5. Q5GenuineYes / no

    Is the message a genuine question or contribution from a person taking part in the conversation?

How it decides

Rules run in code, top to bottom. The first one that matches gives the verdict.

  1. 1
    If Spam score ≥ 70Spam
  2. 2
    If Spam score ≥ 45Maybe spamreview
  3. ·
    DefaultNot spam
Spam score

A 0-100 score weighted from Promotion ×2, Pushes a link ×1.5, Generic ×1.5, Bot patterns ×1, Genuine (inverted) ×2.

Call it from your code

Every template runs as-is with an API key, no setup: POST /api/v1/judges/spam-detector/run.

curl -X POST https://tryevally.com/api/v1/judges/spam-detector/run \  -H "Authorization: Bearer $EVALLY_API_KEY" \  -H "Content-Type: application/json" \  -d '{  "input": {    "message": "🔥 Earn $5,000 a week from home with our trading bot! DM me or join t.me/fastprofits now 🔥"  }}'

Seen in the wild

Community builds on Jev that use this pattern. Not affiliated with Evally.

More in Trust & safety