backservers
where unreel’s bots live

Built for servers that behave more like systems than chat rooms.

This page is not a list of logos. It is a map of the kinds of servers unreel builds for, and how the bots fit into their actual flows: rivals, staff-heavy spaces, internal setups that treat Discord like critical infrastructure.

view bots & systemstalk about your server
moderation-safe flowstickets & incident linesstaff-first interfaces
shape first

Each server gets its own flows. There is no default template it has to squeeze into.

noise down

Bots are designed to stay quiet unless they have something specific and useful to say.

staff first

Staff tools are shaped around how real people use Discord during late incidents.

server archetypes

Not all servers are the same. The bots are shaped for different kinds of gravity.

This is not a catalogue; it is a way to think about the kinds of servers unreel tends to say yes to. Yours might sit between a few of these shapes.

choose the shape
rivals / competitionactive view

These servers need bots to keep match flows, tickets and queue states tight. Players see clean surfaces. Staff see richer context hidden behind controlled entrypoints.

staff surface

How staff see the system on a busy day: panels, commands and quick reads.

core flow

The central loop that the bot helps move forward without adding noise.

state & history

What needs to be remembered between events and how it is kept traceable.

failure modes

What happens when something goes wrong and how staff can see that quickly.

modules

Pieces that slot into the way your server already works.

These are not \"products\" with rigid menus. They are modules that get wired together differently based on what your server is trying to do.

ticketsserver module
Ticket systems that stay in one line.

Bots that keep tickets tidy rather than spraying threads across the server. One clear place to see what is open and who is on it.

core behaviour

Entry, routing and closure flows are designed to match how your staff already think about incidents.

adaptation

No assumptions about categories or layouts. The system adapts to your structure instead of forcing a template.

staff panelsserver module
Staff panels that behave like instruments.

Small surfaces where each action matters. The goal is to make a panel that a new staff member can learn in minutes.

core behaviour

Buttons are chosen based on real incidents, not generic admin tools. If it is rarely used, it might not need a button.

adaptation

Panels are opinionated but adjustable as your team uses them and finds missing controls.

automationserver module
Automation that does not overstep.

Carefully scoped automations that handle repetitive work without making the bot feel unpredictable or overpowered.

core behaviour

The automation is allowed to be smart, but not allowed to be mysterious. Staff should always feel in control.

adaptation

Every automatic action should be explainable after the fact using logs or a simple trace.

signalsserver module
Signals instead of dashboards.

Instead of drowning staff in charts, the bots send focused signals when something crosses a meaningful line.

core behaviour

Signals are contextual, pointing toward the next action rather than asking for a detailed investigation every time.

adaptation

Most of the time, no alert is the best alert. Silence means the system is within expected ranges.

architecture

Under the surface, every server gets a small, opinionated graph.

The exact code is different per project, but the mental model stays predictable: a few nodes passing responsibility between each other.

active node
entry

where the event first touches the bot

In each server, these conceptual nodes are backed by real code and storage, but the mental model stays the same. Events come in, they get routed, state updates, staff see a slim view, and a trace is kept behind the scenes.

The point of this shape is not visual complexity; it is being able to reason about the system on a busy day without opening an editor.

onboarding

Bringing a new server into this kind of system is a narrow, four-step pipeline.

step 1
Start with how the server behaves, not what it could be.

The conversation begins with real channels, real roles and real pain points. No imaginary perfect server, just the one you are actually running.

step 2
Draw the flow from "something happened" to "it is handled."

Together, you map what needs to occur when an incident or request appears. This becomes the backbone of what the bot will move forward.

step 3
Define the surfaces where people will touch the system.

Channels, buttons, panels and messages are decided early so nobody is surprised where the bot shows up.

step 4
Ship a first version and adjust while it is running.

The server gets a working version as soon as it is safe, then the details are tightened while it is under real traffic.

flows

When servers get serious, bots and staff share the work.

The goal is not to replace staff. It is to give them bots that handle repetition and state, while humans handle judgement and context.

flow
what staff do
what the bot does
ticket intake

A member raises an issue that needs staff attention.

staff

Write short responses, close tickets, escalate when needed.

bot

Guide entry, attach context, keep tickets visible and clean.

incident handling

Something breaks, escalates or starts behaving strangely.

staff

Use panels and logs to read what actually happened.

bot

Record transitions, surface signals, keep one source of truth.

queue or match flows

Players or members are waiting for something structured.

staff

Oversee the flow, intervene when needed, adjust rules.

bot

Keep the queue moving, enforce rules, reduce manual juggling.

slow background work

Things that do not need live attention but must be tracked.

staff

Check summaries, trigger occasional manual corrections.

bot

Perform small repeated tasks, log outcomes, avoid surprises.

reliability

Servers that depend on bots need more than animations.

unreel’s work aims to feel premium on the outside, but the parts that matter most are the ones that are invisible until something breaks.

Predictable limits.

Bots are built so you can know up front what they will not do. Clear boundaries are part of the design.

Traceable decisions.

Important actions leave behind enough detail that staff can reconstruct what happened without guessing.

Calm failure modes.

When something fails, it does so in ways that keep your server safe rather than flooding channels with noise.

Human override.

Staff are given ways to override or correct the system without fighting it, especially during incidents.

tiny bleh
blehh

Sometimes a server has a moment that is just pure blehh. The bots cannot stop those. They just make sure the tools are sharp when it passes.

your server
if your server behaves like a system
let the bot carry some of the weight

There is no public list of servers and no broadcast of metrics. If this page feels like the right level of obsessive, the next step is just a single DM.