Moderation

The tools to keep the conversation healthy. They all work in three places: in the dashboard, right inside the widget (for logged-in moderators) and through the management API.

Who can moderate

Who	Where they moderate	What they can do
Site owner	Dashboard, widget and API	Everything, including settings, keys and moderators.
Moderator added in the dashboard	Dashboard and widget	Approve, delete, pin, ban, lock conversations and handle reports. Can't touch settings or keys.
User with `mod: true` in your site's token	Widget	The same moderator actions, right on the page.
Your server, with the secret key	Management API	Everything the owner can do.

To add a moderator, the person needs a Qorum account. In Settings, type their username in Add moderator.

Approval before publishing

In Settings, choose when a comment needs approval before it shows up:

Option	Goes to the queue
No: publish right away	Only comments with a blocked word.
Only comments with a link or image	Comments with a website address or an attached image, plus those with a blocked word.
Everything needs approval	Every comment.

Moderators' comments never go to the queue. Whoever commented sees the notice "Your comment was sent and will show up after moderation approves it".

While it's in the queue, the comment shows up for nobody, doesn't count in the conversation total, creates no notification and fires no webhook. When approved, it shows up with the date it was written, and only then do the total, the notification and the webhook count. When rejected, it's deleted without telling anyone.

The queue is under Moderation → Pending in the dashboard, or at GET /v1/site/comments?status=pending.

Blocked words

One word or phrase per line. If the comment contains any of them, anywhere in the text and ignoring case, it goes to the queue instead of showing up. Careful with short words: "ass" also catches "assistant".

When editing a published comment, the new text can't have a blocked word. The edit is refused with the message "The text has a word that is blocked on this site".

Reports

Any logged-in person can report someone else's comment, choosing a reason. Each person reports the same comment only once. The report shows up under Moderation → Reports, and your server gets the comment.reported webhook.

For each report you can:

Delete the comment: deletes it and marks all its reports as resolved.
Dismiss the report: the comment stays, and its reports leave the list.
Ban the author, together or separately.

Bans

A ban stops the person from commenting, replying and uploading images on the site. They can still read and like. The durations in the dashboard and the widget are 1 day, 7 days, 30 days or forever. Through the API, you choose the hours. Banning again replaces the previous ban.

Banned people see, instead of the comment box: "You are banned from commenting until ... Reason: ...". The time is shown in UTC in English and in Brasília time in Portuguese.

The ban list is under Moderation → Banned, with the button to unban.

Delete

Authors delete their own comments. Moderators delete anyone's. Deleting a main comment deletes its replies too. A deleted comment disappears for everyone, but stays in the dashboard under Moderation → Deleted, where it can be restored.

Locked conversation

Locking a conversation stops new comments on it, without deleting anything. Readers see "Comments on this conversation are closed". Moderators can still comment. Lock and unlock under Moderation → Conversations or with PUT /v1/site/threads/lock.

Pinned comment

Pinned comments show up at the top of the conversation, with the "Pinned" badge. Only main comments can be pinned, never replies.

Moderators pin any comment from the comment's menu. Through the API, choose for how many hours (default 24).
Readers pin their own comment for 24 hours if you grant them the fixar_comentario perk. Use it for subscribers or customers: POST /v1/site/users/:id/entitlements with {"item": "fixar_comentario", "quantity": 1}. Each use spends one unit, and -1 makes it unlimited.

Spam protection

These limits are always on and don't apply to moderators. They sit well above what real people do, so they only stop bots and copy-paste floods:

Limit	Value
Time between comments from the same person	5 seconds
Actions per person (comment, like, edit, report...)	30 per minute
Comments per person	150 per hour and 1000 per day
The same text in the same conversation	Refused for 10 minutes
The same text (25+ characters or with a link) on several pages	Refused from the 4th page within 10 minutes
Image uploads per person	20 every 10 minutes
Requests per IP address	25 per second (bursts up to 120)

Short texts like "thanks!" can repeat across pages: people really do write that on every chapter. A refused comment gets a 429 with a message that says which limit was hit.

Good practices

Start with approval set to "Only comments with a link or image". That's where most spam shows up.
Add moderators instead of sharing your password. Each one logs in with their own account.
Use the comment.reported webhook to notify your team in the channel you already use.
Prefer timed bans. Permanent bans are for repeat cases.