What AI tools can do
Every tool the Brandwise MCP gives connected AI tools, grouped by what it's for.
When you connect an AI tool, it gets the tools below. You don't call them yourself: you ask in plain language, and the AI picks the right tool. Tool names are shown so you can recognize them in your AI tool's activity.
Every connection gets all of these tools. Some use AI credits, see Credits and limits.
Keeping up with new messages
| Tool | What it does | Type |
|---|---|---|
get_ | Lists comments, DMs, email threads, and live chats that are new or changed since the AI tool last checked | Read |
An AI tool that checks your inbox on a schedule can use get_ instead of listing everything again. The first check starts from that moment, and each later check returns only what changed since the one before, with a short preview of each conversation. Changes show up about 10 seconds after they happen.
Comments
| Tool | What it does | Type |
|---|---|---|
search_ | Finds comments that match a question or keywords (up to 40 results) | Read |
list_ | Lists comments, newest first, with filters | Read |
count_ | Counts comments that match filters | Read |
get_ | Opens one comment in full, with its post | Read |
get_ | Counts positive, negative, neutral, and unclassified comments | Read |
cluster_ | Groups comments into themes | Read |
Comments also show whether Brandwise AI moderation has reviewed them, and list_ can show only reviewed or only unreviewed comments.
Moderating comments
| Tool | What it does | Type |
|---|---|---|
moderate_ | Runs Brandwise AI moderation on up to 25 comments and hides what it flags | Write |
hide_ | Hides a comment that Brandwise AI moderation has reviewed | Write |
unhide_ | Unhides a comment | Write |
delete_ | Deletes a comment that Brandwise AI moderation has reviewed | Write |
like_ | Likes a comment as your brand, or removes your like | Write |
The AI tool can only hide or delete a comment after Brandwise AI moderation has reviewed it, either automatically with AI moderation turned on, or on request with moderate_. That way the AI tool can override a moderation decision but not replace moderation. Unhiding and liking work on any comment.
moderate_ reviews comments against your moderation instructions, the same way automatic moderation does. Comments it already reviewed are skipped unless the AI tool asks to review them again, for example after you change your instructions. Your brand's own comments aren't moderated.
These tools only work if the person who connected the AI tool has the Manage comments & DMs permission in Brandwise (owners always do). Hiding, unhiding, deleting, and moderating work on Facebook, Instagram, and TikTok. Liking works on Facebook, TikTok, and LinkedIn. Each action shows in the comment's activity, with the name of the AI tool.
DMs
| Tool | What it does | Type |
|---|---|---|
search_ | Finds DM conversations that match a question or keywords (up to 40 results) | Read |
list_ | Lists recent DM conversations with filters | Read |
count_ | Counts DM conversations that match filters | Read |
get_ | Opens one DM conversation with its messages | Read |
cluster_ | Groups DM conversations into themes | Read |
| Tool | What it does | Type |
|---|---|---|
search_ | Finds email threads that match a question or keywords (up to 40 results) | Read |
list_ | Lists recent email threads with filters | Read |
count_ | Counts email threads that match filters | Read |
get_ | Opens one email thread with its messages | Read |
cluster_ | Groups email threads into themes | Read |
resolve_ | Resolves an email conversation | Write |
bulk_ | Resolves up to 50 email conversations at once | Write |
create_ | Adds an internal note to an email conversation | Write |
Live chat
| Tool | What it does | Type |
|---|---|---|
list_ | Lists recent live chat conversations with filters | Read |
count_ | Counts live chat conversations that match filters | Read |
get_ | Opens one live chat conversation with its messages | Read |
Brandwise AI drafts
When the AI tool opens a comment, DM, email, or live chat, it also sees the draft Brandwise AI wrote for it, if there is one. The list tools include drafts when the AI tool asks for them, and can show only conversations that have a draft. Drafts come with the conversation at no extra cost.
Each draft shows whether it's still waiting, was sent as written, was edited before sending, or wasn't used. For edited drafts, the AI tool also sees what your teammate actually sent. Edit details are kept for 45 days. Each draft also lists the knowledge base pages it used, so the AI tool can spot a recurring mistake and fix the page behind it with apply_.
Your setup
| Tool | What it does | Type |
|---|---|---|
list_ | Lists your connected social accounts | Read |
list_ | Lists your connected email accounts | Read |
list_ | Lists the websites and Shopify stores running live chat | Read |
list_ | Lists your smart labels and the channels each one applies to | Read |
list_ | Lists teammates conversations can be assigned to | Read |
list_ | Summarizes your social and email automations | Read |
list_ | Lists the organizations this connection can reach (only when you connected more than one) | Read |
Organizing the inbox
| Tool | What it does | Type |
|---|---|---|
apply_ | Applies or removes a smart label on a comment, DM conversation, or email thread | Write |
assign_ | Assigns or unassigns a comment, DM conversation, or email thread | Write |
bulk_ | Applies or removes one smart label on up to 50 conversations at once | Write |
bulk_ | Assigns or unassigns up to 50 conversations at once | Write |
create_ | Creates a smart label | Write |
update_ | Changes a smart label's name, description, color, priority, or channels | Write |
delete_ | Deletes a smart label | Write |
Knowledge base
| Tool | What it does | Type |
|---|---|---|
list_ | Lists your knowledge base categories and pages | Read |
get_ | Opens one knowledge base page in full | Read |
apply_ | Creates, edits, moves, renames, and deletes pages and categories | Write |
The built-in Feedback category and its pages can't be renamed, moved, or deleted, though their content can be edited.
Replies
| Tool | What it does | Type |
|---|---|---|
submit_ | Saves a reply draft on a comment, DM, email, or live chat for your team to review | Write |
send_ | Sends a reply right away on a comment, DM, email, or live chat | Write |
bulk_ | Saves up to 50 reply drafts at once, one per conversation | Write |
A draft from submit_ appears in that conversation's reply box as an AI draft. Hover over it to see which AI tool wrote it. A teammate can edit it, send it, or dismiss it. Nothing is sent until someone does.
send_ goes straight to the customer through your connected account. Replies to comments are public unless the AI chooses to reply privately by DM, which works for eligible Facebook and Instagram comments.
Every conversation the AI tool lists or opens says whether it can be replied to right now: whether your team already replied, whether the comment was deleted, and when a reply window closes. Facebook and Instagram only allow DM replies within 24 hours of the customer's last message, and private replies to a comment within 7 days of the comment. This keeps the AI tool from trying replies that would be refused.
The bulk tools work through each conversation separately. If some fail, for example because a conversation was deleted, the others still go through, and the AI tool sees which ones failed and why.
Feature requests
| Tool | What it does | Type |
|---|---|---|
submit_ | Sends your feature request to the Brandwise team | Write |
Ask your AI tool to "send Brandwise a feature request" and say what you'd like. Your AI tool may also send one on its own when a missing tool or option slowed it down. Requests cost no Credits.
Last updated