Jev Moderation Bot
A Discord bot uses Jev to filter spam and scam links and escalate repeat offenses.
Added to Jevfast
How it uses Jev
The bot sends a Discord message with author, account-age, channel, link-indicator, and recent pardoned false-flag precedents as state, then evaluates moderation and profile questions. Results feed progressive warnings or timeouts and moderator review; pardons are saved as future safe precedents.
What Jev decides
User profile requestExample answers · not a recorded Jev response · Source ↗
Question 1 · is scammy
YOUR APP
INSTRUCTION
Does the user show scam, phishing, or fraud behavior?
STATE
The profile state summarizes the member’s messages, account and server age, and prior offenses. These example answers are illustrative.
JEV · NOUL
YesNo
Question 2 · is spammy
YOUR APP
INSTRUCTION
Does the user show repetitive unsolicited promotion or message flooding?
STATE
The profile state summarizes the member’s messages, account and server age, and prior offenses. These example answers are illustrative.
JEV · NOUL
YesNo
App workflow
- Collect message context
The bot records incoming messages and builds a state summary with member and channel context plus saved safe precedents.
- Evaluate the message
Jev classifies the message and profile signals through typed questions.
- Apply progressive action
The bot records offenses and escalates warnings or timeouts according to configured thresholds.
- Review and learn
Moderators can pardon a flag or act from the mod log; pardoned content becomes a future safe precedent.
Why it is interesting
Moderator corrections become server-local examples in later evaluations, connecting human review to the bot’s subsequent context.