Meta Agent
Create the Meta agent
Create the agent that will reply to messages and comments, choosing the right type and its first settings.
With your accounts connected, you create the agent itself. The flow is guided and takes a few minutes: pick the “Meta” type, give it a name and decide how to treat comments.
Open the agent creation flow
Go to “AI Agents”
In yourang.ai's left-hand menu click the AI agents section.
Press “Create agent”
The button is at the top right of the agent list. A guided multi-step window opens.
If the “Meta” card is disabled
In the type-selection screen the “Meta” card stays greyed out and unclickable until the Meta integration is connected: a note under the card invites you to link Instagram and Facebook first. A Meta agent with no connected accounts would have nothing to reply to. Complete the connection (see “Connect Instagram and Facebook”), reload the page, and the card becomes selectable.
The creation steps
Choose the “Meta” type
In the first screen (“Choose the agent type”) you see three cards: Inbound, Outbound and Meta. Click the “Meta” card — the one described as “Replies to Instagram and Facebook DMs and moderates comments”. The selected card is highlighted.
Press “Continue”
You move on to the details screen.
Name the agent
The “Agent name” field is pre-filled with an editable default. Write a recognisable name, for example “Reception” or “Social assistant”. It must be at least 2 characters long.
Decide whether to hide abusive comments
You'll find the “Automatically hide abusive comments” switch (see below). It is on by default: leave it that way if you want the agent to hide spam and insults.
Press “Create agent”
The agent is created and its configuration page opens, where you complete the settings (see “Configure the agent”).
The “Hide abusive comments” option
When it is on, the agent automatically hides clearly offensive or spammy comments instead of replying. Important: genuine questions and legitimate criticism are NEVER hidden — they always get an answer. The goal is to protect your profile's image, not to silence people with something to say. You can change this later from the agent's “Comments” tab, and refine it with rules you write yourself.
Right after creation
Was this page helpful?