Skip to main content
Documentation

AI Control

AI Control is the one place to manage AI for a server. Open the dashboard sidebar and pick AI Control under your server. It brings together your token usage and budget, how replies read, the first-message toggle, the memory review queue, and the results your team is getting.

Pro capability

AI is part of the Pro plan. On Free and Community, AI Control opens in a read-only preview so you can see exactly what Pro does for you. Nothing is saved and the memory queue stays empty until you upgrade.

You need the Manage Settings permission (an admin role) to change anything here. Read-only staff can view the page, but the controls are disabled for them.

Usage and tokens

The usage panel shows how much of your monthly AI token budget you have used and how much is left. AI Assist, the AI flow nodes, and the automation AI reply action all draw from this same pool.

Budget meter

Tokens used this month against your plan's allowance, so you always know where you stand before the cycle resets.

Bring your own key

Add your own AI provider key to run on your account. Usage is still metered and shown, and your own key removes the monthly cap.

Here is the AI token budget each server plan includes per month:

AI budgetFreeCommunityPro
AI tokens / month005,000,000

Configuration

The configuration panel sets how AI-drafted and auto-sent replies read for this server. These defaults apply anywhere the bot writes a reply.

Setting What it does
Reply style Concise, friendly, formal, or apologetic tone for AI replies
Keep replies brief Nudges the model toward shorter answers by default
Reply to first customer message Posts a grounded reply the moment a customer opens a ticket

The Reply to first customer message toggle is the one-click way to answer new tickets. When it is on, a customer's first message gets a reply grounded in your knowledge base, with no flow or rule to build. For more control, build a rule on the On customer's first message trigger instead. See Automations.

Start in draft, then send

If you want a human to approve first replies before they go out, build the reply with an AI Auto Reply in draft mode rather than using the toggle. Drafts post a private, staff-only suggestion with Send, Edit, and Discard buttons. See AI Assist.

Memory review

Memory review is the approval queue for what the assistant is allowed to recall. When a ticket is resolved, its question and answer become a candidate memory entry. Candidates wait here until an admin reviews them, and only approved entries are ever used to ground a reply.

01
Saved on close
A resolved ticket becomes a candidate, at no extra token cost
02
Reviewed here
An admin approves, edits, or rejects each candidate
03
Grounds replies
Only approved entries are retrieved into future answers

Each candidate shows a title, the question, and the answer that closed the ticket. For each one you can:

Approve

Keep the saved answer as it is. From now on it can ground future replies.

Edit

Fix the wording first, then approve. The edited version is what future replies are grounded in.

Reject

Drop it. Rejected entries are kept for the record but never reused.

Per-server and never auto-used

Memory is scoped to your server and is never shared with other servers, and none of it is ever used to train an AI model. Nothing a customer said is reused in a reply until an admin has approved it here. Saving a candidate is a cheap field capture, not a fresh AI write, so building your memory costs you nothing extra.

Results

The results panel shows how AI is performing for your server over the last 30 days, built from real usage and staff feedback. Values with no basis yet read as no data rather than an invented number.

Deflection
How often AI replied in closed tickets
Thumbs-up
Share of AI replies staff rated up
Replies sent
AI replies plus suggestions
Tokens used
Drawn from your budget this month

Staff shape these numbers as they work: a thumbs up or down on an AI reply, or an edit to a draft before sending, shapes which approved answers the assistant reuses. It never trains an AI model, and none of it costs extra tokens. See AI Assist for how the feedback loop works.