What an autoresponder is
An autoresponder is one rule that says: when this happens, do that.
Everything you build with it (a reaction-role menu, an anti-raid gate, a ticket panel, a strike system) is the same four pieces in a different order:
| Piece | Question it answers | Example |
|---|---|---|
| Trigger | When? | someone posts a message |
| Filters | Should we really react? | only in #general, not from staff |
| Actions | Do what? | delete it, warn the author |
| Variables | With which values? | the author's name, the message text |
A rule runs top to bottom, one step at a time. If a step fails, the rest is skipped by default, so a workflow never half-runs without telling you.
The path an event takes
- Discord tells the bot something happened.
- The bot looks up every enabled rule of your server listening for that event.
- Each rule checks its trigger config (does the message actually contain the word you asked for?).
- Then its filters (right channel? right roles? not a bot?).
- Then its gates: the cooldown, and a per-rule rate cap.
- Only then are the actions run, in order, with your variables filled in.
Every fire is recorded. The rule's page shows its recent runs, which step failed and why; you never have to guess whether something ran.
What it is not
It is not a scripting language. There are no loops and no arbitrary code: a rule is a list of steps with optional conditions. That is a deliberate limit, it keeps rules readable by whoever inherits your server, and it makes every rule safe to run on someone else's server, which is what makes the preset marketplace possible.
When you need something genuinely reusable, you do not write a bigger rule: you split it and let one rule emit an event the others listen to.
