Blify Mod

Blify Mod is an AI-powered auto moderation bot for Discord, built on Anthropic's Claude, OpenAI GPT-5, and DeepSeek V4. It's an intelligent automod and anti-spam tool that handles harassment, toxicity, scams, phishing links, NSFW content, and slurs — without the brittleness of keyword filters. Instead of matching banned words, it actually reads the conversation, weighs your server's custom rules, and decides on a proportional action — warn, delete, timeout, kick, or ban. A smarter alternative to traditional word-filter and automod bots for community, gaming, and study servers that want real server safety without false positives.

Why it's different

• Context-aware. When a message looks borderline, Blify Mod pulls in the message it was replying to and the user's prior consecutive messages before deciding. Sarcasm, callbacks, quoted jokes, and "yeah do it" replies don't trip it up.
• Catches what regex misses. Coded harassment, scam/phishing links, NSFW content, spam patterns, and slur variants slip past traditional keyword bots. Blify Mod reads the meaning, not the letters.
• Your rules, only. The bot enforces only the rules you've configured (or the sensible defaults it ships with). 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 worst possible action to a warning, a timeout, or full kicks/bans. You stay in control of the ceiling.
• Reversible by default. Every timeout and ban posts a log entry with a one-click Reverse button, gated by the Moderate Members permission.
• Dry-run mode. Try the bot risk-free for a few days: it logs what it would have done, but takes no action and DMs no one. Perfect for tuning rules before going live.
• Prompt-injection hardened. All user messages are sent inside fenced "untrusted data" blocks, and the system prompt explicitly forbids the model from following any instructions found in them.

Quick setup

1. Invite the bot and move its role near the top of your role list.
2. /enable to switch it on (it's off by default — safe out of the box).
3. /setmodchannels to pick up to 3 channels for active monitoring.
4. /setlogchannel for moderation logs with reverse buttons.
5. /setrules to customize, or use the bundled defaults.
6. Optional: /dryrun mode:on for a few days of safe testing.

Other commands worth knowing

• /immunerole and /immuneuser — mark trusted roles or users as immune.
• /modconfig — see the current configuration at a glance.
• /modlogs — view a per-user audit log of every moderation action, with reasons and reversal history.
• /setappeallink — let timed-out users know where to appeal.
• Or just @mention the bot in any channel for a manual scan.

Tiers

Free tier covers small servers comfortably. Pro and Premium unlock higher channel/rule limits and priority handling. Full pricing on the website.

Ideal for community servers, gaming communities, and study/work servers that want smart server safety and community moderation without false positives.

Privacy

Messages in moderated channels are sent to Anthropic's Claude, OpenAI, and Fireworks AI (which hosts the open-weight DeepSeek model on US-based servers) for evaluation. None of these providers use this data to train models, and Blify Mod has not opted in to any training program. Full details in the privacy docs.

Tags: AI moderation · automod · anti-spam · anti-raid · harassment & toxicity filter · scam & phishing protection · NSFW filter · content moderation · server safety · community management · mod bot