Mokanata Docs

Introduction

What Mokanata is and how its agents work with your team.

Mokanata is an agent-first CRM for revenue teams that sell over messaging channels. Customer conversations from WhatsApp, Messenger and Instagram land in one inbox, with more channels coming soon (these three go live once Mokanata's Meta app is approved). AI agents triage, draft and send replies, enrich records, book meetings, collect payments and watch for churn. People stay in control: they approve, edit or take over anything an agent isn't confident about, or anything a guardrail holds.

How Mokanata works

Connect channels

Link WhatsApp numbers, Facebook Pages and Instagram accounts; more channels are coming soon. Every inbound message becomes a thread on a contact. See Connect a channel.

Deploy agents

Each agent has a role, a model, the channels it works on, the knowledge it's grounded in, and guardrails. You build its behaviour as a visual flow. See Create an agent and Flow builder basics.

Roll out autonomy gradually

Every agent moves through the same three levels: Shadow → Approval → Autonomous.

  • Shadow. It drafts but never sends.
  • Approval. It drafts and a person signs off.
  • Autonomous. It sends on its own unless a rule holds the draft.

See Deploy, pause and set autonomy.

Review by exception

Drafts that are low-confidence, escalated or held by a guardrail go to the approval queue. Each one carries its confidence score, its cited sources and its reasoning. Auto-approve rules release the routine ones. Guardrails can never be overridden by a rule. See Triage the approval queue.

Measure

Reports cover conversations, SLA, AI vs team performance, campaigns, payments and pipeline. Each agent also has its own runs and trends. See Reports.

Find your way through the guide

The guides follow the dashboard's sidebar. New to Mokanata? Start with Create your account and Set up your workspace, then connect your first channel.

On this page