Blify Mod Documentation

Blify Mod is an AI-powered Discord moderation bot built on Anthropic's Claude. Instead of relying on keyword lists, it reads the surrounding conversation, weighs your custom rules, and decides on the appropriate action — warn, delete, timeout, kick, or ban — all with a one-click reversal button in your log channel.

These docs cover everything you need to get the bot running, configure it for your server, and understand how it makes decisions.

#Quick links

• Getting started — invite the bot, enable it, and run your first scan.
• Setup guide — pick channels, set a log channel, customize rules.
• How it works — the screen → judge → reconsider pipeline, in plain English.
• Tiers and pricing — Free, Pro, and Premium feature limits.
• Privacy and data — what data is sent where, and what isn't.

#What makes Blify Mod different

• Context-aware judgment. When a message looks borderline, the bot pulls the message it was replying to and the user's prior consecutive messages before deciding — sarcasm, callbacks, and "yeah do it" replies don't trip it up.
• Your rules, only. The bot will only enforce rules you've configured (or the bundled defaults). It won't invent rules from "common sense" or Discord's ToS — if you haven't listed it, it isn't moderated.
• Strictness caps. Lock the bot's worst possible action to a warning, a timeout, or full kicks/bans — your call.
• Reversible by default. Every timeout and ban posts a log entry with a one-click Reverse button (gated by Moderate Members).
• Dry-run mode. Try the bot for a few days with zero side effects: it logs what it would have done, but takes no action and DMs no one.
• Prompt-injection hardened. User messages are sent inside fenced "untrusted data" blocks, and the system prompt explicitly forbids the model from following any instructions found inside them.

#Need help?

The fastest way to get a real human is the support server. Bugs, feature requests, and abuse reports all go there.