Mamori

守knowledge, guarded

Answers from your documents.Nothing invented.

Mamori reads the files you give it and answers your members inside Discord. When nothing you have written down covers the question, it can read the channels those members can already see, say where the answer came from, and log the gap. No guessing, and no bill you did not agree to.

  • Answers from your documents from $5 a month
  • Free plan runs the moderation side, no card
  • Hard monthly cap, no overage billing

On a loop: one reply straight from a document, one where nothing was written down so Mamori read the channels those members can already see. Never the ones they cannot.

What it does

Three things carry it.

Three things carry the product: it answers from your documents, it shapes a server in one command, and it never spends more than the plan you picked. Everything else is a module you switch on.

A knowledge base that refuses to guess

Upload markdown, text, HTML or PDF. Mamori retrieves the closest passages and answers from those alone. When nothing matches closely enough it says so and files the question for you, or reads the open channels for you if you asked it to.

One command to shape a server

Run setup, pick a template, and Mamori shows the exact channels, roles and permissions it would create. Nothing is applied until you confirm the preview.

Spending you can actually cap

Every plan buys a fixed number of AI replies per month. There is no overage billing and no metered surprise. At the limit Mamori stops answering and everything else keeps running.

Modules

Everything else is a switch.

26 modules on the dashboard, 11 of them on the free plan. Each one is off until you turn it on, each one keeps its settings when you turn it off, and none of them punish anybody until you pick an action that does.

The raid shield and anti nuke both stand down for the server owner, for anybody above Mamori in the role list, and for the roles you mark as exempt. A lockdown writes down the old permissions first and puts them back when the window ends, so a missed run never leaves a channel shut.

How it works

Live in about ten minutes.

01Invite

Choose where it may speak

Add Mamori, then name the channels it is allowed to answer in. It stays quiet everywhere else.

02Upload

Feed it your documents

Drop files into collections. Mamori splits them into passages and indexes them for retrieval.

03Ask

Members ask, Mamori cites

Answers arrive with the document they came from, so anyone can check the source in a second.

For admins

Help for whoever runs the place.

Members get answers from your documents. Admins get something separate: an assistant that reads this server's own numbers and writes the messages nobody enjoys writing.

Ask about this server, not Discord in general

Why nobody is opening tickets, what went unanswered last week, what to fix first. It answers from your own numbers and names the command to run, or says plainly that it cannot see the answer.

Announcements written for you

Describe the message and it comes back as a finished embed, with headings, bullets and numbered steps where the order matters. You read it before anyone else does.

Edits without retyping

Point it at a message Mamori posted and say what should change. It returns the whole thing rewritten, and anything you did not mention comes back word for word.

A straight read on the server

Members, tickets, unanswered questions, moderation, quota and the settings still switched off, followed by the three fixes worth doing first.

It needs the Administrator permission in Discord and a plan from $15 a month. Nothing reaches a channel on its own: every draft is shown to you first, and it is posted only when you press the button.

010203

Each arc is one limit. A request has to clear all three before a single token is spent.

Cost safety

Three limits, all of them hard.

Runaway model bills are why bots like this quietly disappear. Mamori treats spend as a product constraint rather than something to watch nervously in a dashboard.

01

Per-server monthly quota

per plan

Each plan buys a fixed number of AI replies. At the limit Mamori stops answering until the 1st. It cannot spend past the number you picked.

02

Per-user daily cap

25 / day

Every member gets the same small daily allowance, plus a cooldown between requests, so one person cannot drain a whole month in an afternoon.

03

Semantic cache

0.94 match

Repeat questions are matched against answers already given. A close enough match is served from storage and costs nothing to run.

Everything that is not AI stays available whatever happens to the budget, including on the free plan.

What it costs

Free is real, but it does not answer.

The free plan runs the moderation side of Mamori: server setup, tickets, automod, role panels and 11 of the 26 modules. Answering from your documents is the paid half, and it starts at $5 a month.

Free$0, no card

No answers from documents. Setup, tickets, automod, role panels and the free modules all work.

Starter$5 / mo

500 AI answers a month, 25 pages of knowledge base, 1 server.

Pro$15 / mo

1,500 AI answers a month, 500 pages of knowledge base, 1 server.

Network$40 / mo

6,000 AI answers a month, unlimited pages of knowledge base, 10 servers.

Every plan is a ceiling rather than a starting point. At the limit Mamori stops answering until the 1st, and nothing bills you for going over, because you cannot go over. Full comparison.

Put it in your server and see what it does not know.

Install it free and run the moderation side for as long as you like. Upgrade when you want it answering from your own documents.