The AI auto‑reply inbox for social DMs

All your social DMs in one inbox. AI answers the common questions, you handle the rest.

Thanks! We'll be in touch.

Uses your knowledge base
Inbound DMs auto triage
Manual or auto send

Watch a quick demo

See the setup, triage, and the approval window in ~2 minutes.

2 min demo • shows setup + reply flow

How it works

Connect accounts, add knowledge, then review queued outgoings before they send.

1

Connect your social accounts

Link the inboxes you manage so DMs arrive in one place (X today; Instagram, Facebook, WhatsApp, and Reddit next).

2

Connect a knowledge base

Upload FAQs, policies, and product info. Replies are grounded in what you provide.

3

AI triages every message

Each inbound DM is triaged against your knowledge base. This is a fast estimation of whether the topic is covered in the KB. High-confidence matches are marked “can auto-reply”.

4

AI agent drafts grounded replies

The AI agent goes deeper on each topic, analyzing your KB entries in detail to craft accurate replies. No generic responses, no hallucinations, only answers grounded in what you provided.

5

Review before it sends

Drafted replies are scheduled for sending in 5 to 60 minutes. Skim through the outgoings, edit the drafts you don't like, or stop them from sending altogether.

Features that save you hours every week

Unified Inbox

All your DMs in one place

Stop switching between X, Instagram, Facebook, and WhatsApp. See all your social DMs in a single unified inbox. Respond faster, never miss a message, and keep your team on the same page.

  • Connect multiple accounts from each platform
  • Real-time messages
  • Filter by platform, urgency, or triage score
Unified inbox showing DMs from multiple platforms
AI-Powered

AI triage scores every message

Every inbound DM is triaged against your knowledge base. This is a fast estimation of whether the topic is covered in the KB. High-confidence matches are marked “can auto-reply”, then an AI agent goes deeper to craft accurate, grounded responses.

  • Semantic matching understands intent, not just keywords
  • Replies are grounded in your KB, no hallucinations
  • Two-stage validation ensures accuracy and safety
AI triage showing match scores and suggested replies
Safe Auto-Reply

Auto-reply with a review window

Messages marked “can auto-reply” get a drafted response queued for sending. But nothing goes out immediately, replies sit in a review window so you can skim, edit, or cancel before they reach the sender. Enable or disable auto-reply at any time with one click.

  • Configurable review window (5-60 minutes)
  • One-click cancel or edit before sending
  • Full audit trail of what was sent and why
Review queue showing pending auto-replies with timers

Built for safe DM automation

A lightweight “mini agent” for social teams, knowledge‑base driven, confidence‑gated, and easy to control.

Semantic matching

Understands intent, not exact keywords, so “shipping times?” can match “delivery timeline” in your docs.

Auto‑reply, safely

When the AI is confident and grounded, it drafts from your KB and queues an outgoing reply, no guessing.

Review window

Outgoings sit in a timed queue so you can skim, stop, or edit before they send.

Easy KB setup

Your knowledge base can be in any format, just upload documents like product descriptions, FAQs, or policies. No rigid structure required.

Simple controls

A single threshold to set, to decide what is eligible to auto‑reply. Turn on auto‑send or work manually with KB‑suggested drafts.

Audit trail

Every action is logged (match score, source, sent vs. deferred) so you can prove what the bot did and why.

Ready to automate simple DMs safely?

Sign in, connect X, paste a KB, and start reviewing outgoings. Turn on auto‑reply when you’re ready.