Sales-ops manager

Customer-Ops

In development

Every customer's invoices, emails, meetings, and projects in one live 360, with an AI Relations Officer who drafts the next move.

Product proof

Product images are being prepared.

The workflow below is sourced from the product record.

Access

Live in production behind single sign-on. We run Magic Unicorn's own customer pipeline on it every day. Public onboarding and pricing are coming; follow the build log for the opening.

The work

What it does

  1. Builds a live customer-360 by federating contacts, invoices, email threads, meetings, and projects from Contact-Ops, Accounting-Ops, Email-Ops, Meeting-Ops, and Project-Ops. It reads live, so there is no second database to keep in sync.
  2. Runs five relationship agents on staff: lead triage, next-best-action, meeting prep, follow-up drafting, and collections nudges. They run on local models at $0 inference, with per-call opt-in to Claude.
  3. Answers pipeline questions in chat, and in Operator mode the Relations Officer proposes writes (stage moves, notes, mark-lost, reopen) behind a propose, confirm, execute gate. Every action lands in an append-only audit log.

Watch · 1:41

Watch the Customer-Ops walkthrough

See Customer-Ops handle the work in a short, practical walkthrough.

The workflow

How it works

  1. 01

    Sign in with your Unicorn Commander account. Customer-Ops pulls your contacts, books, mail, meetings, and projects live from the apps that own them.

  2. 02

    The Relations Officer and five relationship agents watch the pipeline, triage new leads, and draft the next move for every deal.

  3. 03

    You confirm proposed actions with one click. Everything executes with full audit trail behind it.

Source record

A claim with a source

Live in production at customer-ops.unicorncommander.ai since June 2026, running Magic Unicorn's own customer pipeline every day (13 real relationships and counting).

Keep exploring

All AI employees

For humans and for agents

Bring your own agent.

Unicorn Commander is built for people and agents. Start with the connection details and the permissions for the product you want to use.

Machine discovery

/.well-known/agent-card.json

The card describes the available identity and capabilities. It does not grant anonymous access.

A useful next step

Need a guided fit?

Start with the job, the product state, and the way you want to run it.

Book a call