Start here
Getting started
Invite AETHON, configure it and run your first request.
This guide takes a server administrator from the invitation to a first verified change.
1. Invite AETHON#
Use the Invite Bot button on this website and choose your server (Discord only lets members with the Manage Server permission add bots). The invitation asks for the permissions AETHON needs and never for Administrator.
- Keep every requested permission. Before acting, AETHON checks its own permissions and refuses operations that need a missing one; the message names the missing permissions.
- Move AETHON's role above the roles it should manage. Discord only lets a bot manage roles below its highest role, and never roles managed by an integration.
- Keep your staff roles above AETHON if you want the role hierarchy to protect them.
2. Open the setup panel#
Run /setup panel (AETHON level admin; members with the Discord Administrator permission are admins under the
default policy). The panel saves each setting immediately and records it in the audit trail:
| Setting | What it does |
|---|---|
| Language | Server language for AETHON's messages, including private replies, and slash command descriptions and choice labels. English by default; an administrator can change it with /settings language. |
| Progress channel | Where the single progress message of each project is posted and edited. |
| Approval channel | Where approval cards are posted. |
| Approver role | Role pinged when an approval is needed. |
| Alert channel | Moderation log and alerts. |
/settings view shows the current values. AETHON needs View Channel and Send Messages in the progress and approval
channels.
3. Send your first request#
/agent request: Create a support area with tickets and a Support role dry_run: true
- AETHON inspects the server, plans the work and validates the plan before anything changes.
- With
dry_run: trueit stops at a preview. Press Execute (or use/projects execute) to run it: the server is re-inspected and the plan re-validated first. - If something is unclear, AETHON asks a question; answer with the Reply button.
- The progress of the project is one message that is edited in place.
You can also start without the agent: /setup template builds a complete structure from a ready-made blueprint —
programming, gaming, study, support, content creator or general community, ticket system, welcome & verification, or
moderation baseline. Add dry_run: true to preview it.
4. Approve sensitive operations#
Under the default policy LOW and MEDIUM risk operations run automatically when the requester's level allows them, and HIGH and CRITICAL operations wait for an approval: HIGH needs an AETHON admin, CRITICAL the server owner. Some operations (deleting channels or roles, bans, project rollback) always need an approval and can never run inside an automation.
Approve or reject from the approval card, or with /approve list and /approve decide. An approval only covers the
exact operation and the state of the resources it was requested for; if either changes, AETHON asks again. Approvals
expire if nobody decides them.
5. Follow up#
| Command | Use it to |
|---|---|
/projects list, /projects view | follow projects and read their final report |
/tasks list, /tasks view | inspect individual tasks |
/cancel | cancel a project |
/server dashboard | see the state of the server |
/audit history | review everything AETHON changed |
/usage | check credits, write operations and plan limits |
/help | open the built-in help topics |
Operator levels#
| Level | Who has it | Runs directly | Can request with approval |
|---|---|---|---|
| viewer | members granted it with /permissions grant | LOW | LOW |
| operator | granted, or members with Manage Server (default policy) | up to MEDIUM | up to HIGH |
| admin | granted, or members with Administrator (default policy) | up to HIGH | up to CRITICAL |
| owner | the server owner | everything | everything |
/permissions level shows your level, /permissions operators lists operators, and the server owner can change the
policy per risk level (/permissions risk) or per tool (/permissions tool). Continue with
Agent & projects, Approvals & security and the command reference.