Jev Moderation Bot
Discord bot where Jev flags spam and scam links in real time, drives an escalating timeout ladder and profiles members.
Overview
Jev Moderation Bot is a Python Discord bot that uses TypeSafe's Jev to moderate messages as they arrive. Messages flagged as spam or scam links are deleted with a reason in the audit log, and offences escalate: a warning, a final warning, a 10-minute timeout and then 1-hour timeouts, with durations configurable.
Moderators can run /profile on a member to score their recent messages for scam risk, spamminess, noobness, toxicity and helpfulness and assign a persona such as beginner, regular, spammer or troll. Alerts go to a mod-log channel with one-click pardon or ban buttons. When an admin pardons a message, it is saved as a safe precedent and included in future checks for that server, reducing repeat false alarms. Messages are cached in SQLite, confidence thresholds are configurable, and flagged messages and pardons can be exported for review.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Is this message spam or a scam link?(inferred) | yes · no |
| Score | How risky is this member (scam risk, spamminess, toxicity, helpfulness)?(inferred) | — |
| Choice | Which persona fits this member? | beginner · regular · spammer · troll |
How it works
- Each incoming message is evaluated by Jev with server precedents
- Flagged messages are deleted and logged
- Offence count drives warning or timeout
- Moderators review alerts and can pardon, which saves a safe precedent
Strengths
- Real-time deletion with an audit trail
- Pardons become precedents that reduce false positives
- Configurable thresholds and timeouts
- Export of flags and pardons for offline review
Caveats
- Automatic deletion can remove legitimate messages before review
- Member profiling raises privacy considerations
- Requires a TypeSafe key and Discord bot token
Getting started
With Python 3.10+, a Discord bot token and a TypeSafe key, follow the README setup and run the bot.At a glance
- Type
- Project
- Status
- beta
- Language
- Python
- Runtime
- Discord bot
- Has code
- yes
Use cases and integrations
How curated lists describe it
Community moderation: Discord bot that scores incoming messages for phishing, spam, and social engineering with Jev and drives a four-stage escalation ladder, injecting pardoned messages back into context as verified-safe precedent.
yibie/awesome-jevA Discord moderation bot using Jev to check spam and scam links, with local warning and timeout rules.
logicrw/awesome-jev-projectsReal-time Discord moderation bot: Jev evaluates messages and metadata in parallel to catch phishing, spam, and social engineering with a progressive escalation ladder.
hellogumbo/awesome-jevDiscord bot that filters spam and scam links in real time and escalates repeat offenses.
MrJev/awesome-jev
Frequently asked questions
What does Jev decide?
Whether a message is spam or a scam link, plus profile scores and a persona for members on request.
How does it reduce false positives?
Pardoned messages are saved as safe precedents and included in future checks for that server.
What happens to repeat offenders?
Warnings, then a 10-minute timeout, then 1-hour timeouts, all configurable.
Related projects
Chrome extension that uses your Jev key to hide high-confidence sexual and solicitation replies on X behind reversible placeholders.
Browser extensionproductionTypeScriptWordPress plugin that triages comments in the background with Jev questions on relevance, spam and abuse.
ProjectbetaPHPMessage moderation that asks Jev for probabilities of spam, scam, harassment and other harms plus your own plain-English rules.
ProjectbetaPythonChrome extension and Workers API that score X posts and replies for AI writing, solicitation and paraphrase bots using Jev.
Browser extensionbetaTypeScriptChrome extension that labels every X post you scroll past with six typed Jev questions and shows a live cost counter.
Browser extensionbetaTypeScriptScam text checker where code extracts link and sender facts, Jev answers eleven narrow questions, and a pure policy scores risk.
ProjectprototypeTypeScript