Skip to content
All Jev use cases

Prompt pack

Jev content moderation prompts

Yes or no checks for user comments and messages: toxicity, insults, identity attacks, threats, sexual content, profanity and spam, plus one request that runs five checks at once. Every Jev content moderation prompt was run on real Jev against rater labels from public datasets, with the agreement rate shown for each.

Jev content moderation is a set of yes or no checks for comments and text messages. It can flag toxicity, insults, identity attacks, threats, sexual content, profanity, or spam. One paid prompt runs five of those checks in one request. Use Jev content moderation as a first pass before a comment stays up or a message is dropped.

The weak line in Jev content moderation is Run five moderation checks in one request: agreement was 77% on 50 examples. If you need one label, run that check and send the middle to a person. Agreement is the share of examples where Jev content moderation matched the human label. If unsure, hold the comment for a person. Before you hide a comment or drop a message, run Jev content moderation on your own comments. This page is unofficial. TypeSafe does not run it. Jev content moderation here is the prompt and the test result. You use your own key to call.

How we tested these Jev content moderation prompts

Model
typesafe/jev-1.14-20261014
Tested on
Sent as
Decisions API request (state + questions) → typesafe/jev-1.14, served by TypeSafe

We ran these tests ourselves. TheJevAI is an unofficial, independent site and is not affiliated with TypeSafe.

Free Jev content moderation prompts

Copy a request, put your own data where the placeholders in double braces are, and send it. To send it to TypeSafe's API, add "model": "jev-1.14.0" to the request body.

Download the free prompts (JSON)

Jev content moderation: Flag a toxic comment

This Jev content moderation check says whether a comment is rude or disrespectful enough to drive people out of a discussion.

Request body
{
  "state": {
    "comment": "{{comment}}"
  },
  "questions": {
    "flagged": {
      "type": "noul",
      "instructions": "Is `comment` toxic, meaning rude, disrespectful or unreasonable in a way that would make someone want to leave the discussion?"
    }
  }
}

Put the comment in `comment`. `noul` is the probability that raters would call it toxic. Jev caught all 20 toxic comments but also flagged 8 of 20 that most raters let through, often blunt or sarcastic political remarks. If false alarms are costly, raise the cut-off or send the middle band, roughly 0.5 to 0.8, to a person. What counts as toxic depends on the community, so try it on a sample of your own comments first.

80% agreement (32 of 40): toxicity rater majority, Jev content moderation yes at 0.5 or more

Dataset: Civil Comments (Jigsaw)License: CC0 1.0

Jev content moderation: Flag a spam text message

Use this Jev content moderation check to say whether a text message is bulk spam, such as an ad, a prize scam or a premium-rate lure.

98% agreement (39 of 40): hand labeled spam or ham, Jev content moderation yes from 0.5Dataset: SMS Spam Collection (UCI)

Jev content moderation: Flag an identity attack

Use Jev content moderation to say whether a comment attacks people for their race, religion, gender, sexuality or a similar trait.

85% agreement (34 of 40): identity attack majority, Jev content moderation yes at 0.5 or moreDataset: Civil Comments (Jigsaw)

Jev content moderation FAQ