Question 1

How is ModCheck different from AutoMod?

Accepted Answer

AutoMod uses machine learning to catch hate speech, profanity and harassment, however, it doesn't understand stream context. It'll flag "did you kill the dinosaur?" as violent when someone's playing Tomb Raider. Or block "take your shirt off" when a character needs to remove armor to heal. ModCheck understands what's actually happening in the stream - the game, the streamer's mood, what they are saying and chat history. Context is included by design. Judges are plug-and-play, fully customizable and get smarter over time as they learn from your community.

Question 2

How is ModCheck different from Nightbot or Sery_bot?

Accepted Answer

Nightbot handles spam with rule-based filters - caps limits, link blocking, blacklisted words and regex. Sery_bot protects against hate raids, follow bots and scam accounts. Both are great at what they do. ModCheck solves a different problem - understanding what messages actually mean. It uses Judges written in plain English that factor in the game being played, the streamer's mood and chat history to catch things like backseat gaming, passive aggression and pile-ons. They work well together. Use Nightbot and Sery_bot for spam and bots, ModCheck for everything that needs context. Next up, I want to make it easy to integrate ModCheck with existing bots so you can have the best of both worlds without juggling multiple dashboards.

Question 3

How does ModCheck work under the hood?

Accepted Answer

ModCheck builds a live picture of your stream - who you are, what you're playing, what's been said and what's happening right now. Every message gets checked against that picture. The first pass is fast - if ModCheck has seen something similar before in this context, it already knows the answer. Think of it as memory. The second pass is a finetuned AI model tailored to each Judge, which handles anything new or ambiguous. The result is that your Judges get faster and sharper the more you stream. Two streamers with the same rule will get different results because the system adapts to each community. More detailed technical docs are on the way.

Question 4

How fast is ModCheck?

Accepted Answer

Messages are judged in under 500ms. Fast enough for the biggest chats on Twitch. Judges run in parallel, so adding more rules doesn't slow things down. The more judges are used the faster they become, as they learn from more messages in your chat history.

Question 5

Can I customize what gets flagged?

Accepted Answer

That's the whole point. You write Judges in plain English - like "flag passive-aggressive comments when I'm on a loss streak" or "block spoilers for the game I'm currently playing blind."

Question 6

How much does it cost?

Accepted Answer

Three self-serve plans: Starter at $14/month (3 judges), Pro at $39/month (6 judges) and Partner at $119/month (12 judges), plus a contact-us Enterprise tier for the biggest channels. The difference is how many judges run at once and how many messages get judged each month - every plan has the full feature set. Every plan starts with a 3-day free trial. You add a card up front, we charge nothing until the trial ends, and cancelling is one click from your channel page.

Capability	ModCheck	AutoMod	Nightbot	Fossabot	Sery_Bot
Context-aware behaviour judgement	Yes	No	No	No	No
Plain-English custom rules	Yes	No	No	No	No
Per-message reason + confidence	Yes	No	No	No	No
Shadow-test & VOD replay	Yes	No	No	No	No
Human-in-the-loop review	Yes	Yes	No	No	No
Slurs & banned words	Yes	Yes	Partial	Partial	Partial
Spam, caps & link filtering	Yes	Partial	Yes	Yes	Yes
Hate-raid & follow-bot protection	No	Partial	No	Partial	Yes
Commands, timers & loyalty	No	No	Yes	Yes	No

How ModCheck compares

Head to head

Your community
deserves better.

How ModCheck compares

Head to head

Your communitydeserves better.

Your community
deserves better.