Help

The first two sections are all most people need. Everything after them is optional.

Protect or analyze a post

Sign in with X, Google or Microsoft. A team or agency gets one workspace: connect each client's X account to it (the client approves through X, so no passwords are shared) and invite your staff.
Paste a post's link in the box at the top of your dashboard and choose:
- Protect, for one of your own posts. Replies already under it (up to 7 days back) are read first, then new replies and quote posts every few minutes for 7 days. Abuse is hidden and the accounts that sent it are handled as you set under Automatic actions. The replies are also grouped into themes.
- Analyze, for any post, yours or anyone's. Its replies and quote posts are read for 7 days and grouped into themes with summaries. Nothing is hidden or muted, so you need no rights over the post.
Each post you protect or analyze gets a card on your dashboard, with a short title, what it says and its numbers. Click it to open that post's own page.

After your first sign-in you can also start protection without leaving X: reply @mydarboon to any post of yours and it is protected within a few minutes. It works on your replies in other people's threads too, though there Darboon can only find the attackers, not hide their replies.

Every new post, automatically: under Settings on your account page, switch on "Protect every new post automatically" for an X account, and each post it makes from then on is protected as soon as it appears, with your saved themes that go on every post. It is set per X account, so in a workspace each brand or client account is switched on or off by itself.

Trial: your first $5 of usage needs no card. Stop within it and you pay nothing; carry on with a card and the $5 is billed with your first charge. After that you pay as you go, and every post has a spending cap, so a flood of replies pauses the post instead of running up a bill. With gift credit from someone, no card is needed until it runs out.

A post's page and its themes

A post's page shows how many replies and quotes were read, how many accounts were flagged and hidden, and what it has cost so far. Below that, What the replies are about lists its themes, and then the accounts flagged on the post: newest first, more loading as you scroll, new ones appearing at the top by themselves.

Themes are found by grouping the replies and quote posts by what they say, in English, Persian (and Finglish) or the language of the post, as they arrive. Nobody sets up keywords: a post about a crash might get "Autopilot blamed for the crash" and "Refund demands". Each reply is then read against every theme of the post and placed in all the themes it fits, so a reply that blames the crash and asks for a refund is in both. Text inside pictures (memes, screenshots) counts with the reply it came in, when image checking is on. A post is first grouped once about 30 replies with words have been read, and the themes are kept up to date after that.

Themes of your own: under the themes, "Add a theme of your own" takes a description in plain words, such as "insults the author", "asks about the price" or "interested in the new colours". Every reply that fits is collected under it, the ones so far and new ones, and a reply can be in several themes at once. A post can have up to 5 of your own; checking costs 0.1¢ per reply, however many there are.

Each theme has a title, a summary written for you (and rewritten as the theme grows), example replies, a CSV download of its replies, and three switches:

Hide replies: its replies, so far and from now on, are hidden under your post.
Mute accounts: the accounts that wrote them are muted for you.
Send to a list: the accounts are written to a Google Sheet you choose, for your records, your team or your client (see Lists). This alone hides and mutes nothing.

Nothing happens to a theme's replies until you switch something on. On a post you only analyze, adding the accounts to a list is the one choice. The "ⓘ What these do" link under the buttons explains each one.

What Darboon can and cannot see

You sign in through X itself, and before you approve, X shows you exactly what Darboon is asking for.

It can: see your X handle and name, read replies to posts, hide replies under your posts and mute accounts for you.
It cannot: see your password, your email address, your phone number or your direct messages. It cannot post, like, follow or send messages as you.
It learns nothing about who you are beyond the X account you sign in with.
You can cut it off at any moment in X under Settings > Security and account access > Connected apps.

How strict it is, and who is left alone

On the dashboard, Protection level has three settings. Lenient acts only on clear abuse such as slurs, threats and profanity. Standard also catches insults and insulting emojis. Strict removes anything rude, mocking or dismissive and keeps only supportive, neutral and politely disagreeing replies. A new level applies to replies checked from then on; "Rescan all replies" on a post's page re-checks what is already there.

People you follow are always left alone, whatever they say. If you follow someone Darboon had already flagged, that is undone automatically, as if you had clicked "Not hostile, undo".
Arguments between other people are not pinned on you: a reply answering someone else is judged only by what it says about you. An attack on someone you follow counts as an attack on you.
Your own accounts count as you: the X accounts in your workspace are never flagged on each other's posts.

If Darboon gets one wrong, click "Not hostile, undo" on that account. That one button reverses everything: its replies are unhidden, it is unmuted, it is taken off your lists, and it is never flagged again.

What it does, and what that costs

Under Automatic actions you choose whether Darboon hides hostile replies and whether it also mutes their authors. Reading replies and each hide or mute are charged as you go; the rates are shown next to each switch.

Hiding removes a reply from the thread under your post. X still shows a hidden reply to anyone who opens its own link, and it can still appear in people's timelines; muting is what keeps an account out of your own feed.

Checking images is an extra you can switch on in the same place. With it on, Darboon also looks at the pictures attached to replies (up to two per reply; for a video or GIF, the still frame) and reads the text inside them. It is charged per picture checked, and a picture Darboon has checked before at your protection level, on any post and for anyone, is not charged again. A picture is fetched, read and discarded; only a fingerprint of it, the text read from it and the verdict are kept, never the picture, and nothing about who posted it or where.

Gift protection to someone

You can pay for someone else's protection, for example a friend who is being attacked. A gift is $5 of protection, enough for roughly 180 replies on a typical post. It goes on your tab like your own usage.

There are two ways to give one:

On the site: on your dashboard, under "Gift protection", type their X handle and click "Send gift". Darboon's X account sends them a direct message about it if their settings allow, and you can also announce it with the "Announce it in a post" link, which opens a ready-made post from your own account.
On X: reply @mydarboon gift (or @mydarboon هدیه) under one of their posts. The bot replies in that thread, in the language of the post, saying that the post's author has been given a gift and how to use it, so they see it under their own post.

You can add a message, such as who you are or why you are doing this: in the box next to the handle, or after the word gift in the mention (@mydarboon gift from your neighbour, hang in there). Darboon sends it with its message about the gift. Links are taken out, and a message that reads as unfriendly is not passed on.

What the person you gift needs to know:

A gift pays; it does not switch anything on. Only the author of a post can ask for it to be protected, so nobody can turn on hiding and muting for someone else's post.
They sign in with X once, which gives Darboon permission to act on their posts. They need no card.
They then protect any post of theirs, on the site or by replying @mydarboon under it. The gift is used before anything of their own, and if several people gift them, the oldest gift is used first.
When the gift runs out their post pauses, unless they have added a card of their own.

A gift to someone who never signs in lapses after 2 days, and its value comes back to the person who gave it, as credit. Once they have signed in, it does not expire. You need a card on file to send one.

Lists in Google Sheets

You do not need this to use Darboon. A list is a Google Sheet that Darboon writes accounts into, one row each: the account's id, its handle, the date, your handle, a link to the reply, and the theme it came from. Use one to keep a record, to share with your team, or to hand to a client. Give each list a name on your dashboard so you know which is which.

Each list either receives every account flagged as hostile on your posts, or only the themes you send to it ("Send to a list" on a theme). You choose this per list, and can change it.

The easy way: let Darboon create the sheet

On your dashboard, under Your Google Sheets, type a name for the list (for example "Block list" or "Refund demands") and click Create a Google Sheet for me.
Google asks you to let Darboon create the sheet in your Google Drive. Darboon can then open only the sheets it creates for you, nothing else in your Drive. You can remove that access in your Google account at any time.
The sheet appears in your Google Drive and on your dashboard. It is yours: share it from Google Sheets like any other sheet if you want others to see it.

Or use a sheet you already have

Open the sheet in Google Sheets, click Share and add this address as Editor (untick "Notify people"). It is Darboon's own Google identity, which it writes with:
In the sheet's first row, in any cell from column G onward, type your X handle: @your_handle. Only people who can edit the sheet can do this, so it tells Darboon that the sheet's editors allow your account to write to it. Several Darboon users can share one sheet: put each handle in the first row.
On your dashboard, under Your Google Sheets, paste the sheet's link, give it a name and click Connect this sheet.