Human Agent Tag

The Human Agent Tag (HUMAN_AGENT) is a permitted message tag in Meta's Messenger Platform that extends the standard 24-hour messaging window to seven days, on the strict condition that a real human — not an automated workflow — composes and sends the reply. Meta introduced it for support workflows where complex issues legitimately can't be closed within a day.

Technically you attach the tag in the send call: {"messaging_type": "MESSAGE_TAG", "tag": "HUMAN_AGENT"}. The 24-hour clock is bypassed and replaced by a 168-hour clock that resets on every user reply. Once the seven days lapse, the only remaining options are message tags (POST_PURCHASE_UPDATE, ACCOUNT_UPDATE, CONFIRMED_EVENT_UPDATE) and paid Sponsored Messages.

Why it matters — and the trap

The tag is widely misused by automation tools as a backdoor extension to bot-driven flows. Meta's policy is unambiguous: automated replies are prohibited under the Human Agent tag. The bot platform itself cannot claim it. A human teammate has to be inside the inbox view, typing or pasting the reply, for the tag to be lawful. Tools that quietly attach the tag to scheduled or rule-triggered messages risk an app-level downgrade or revoked access.

Gotchas

• Sending automated content with the tag attached is the single most common reason platforms get throttled or moved down a messaging tier.
• The tag does not apply to first contact — you still need a prior user-initiated thread or a valid private reply to engage.
• It is only available to apps with Advanced Access to pages_messaging or instagram_business_manage_messages.

For the operational rules a team should put in place before exposing the tag in product, see Instagram DM auto-reply rate limits.