Description
How Mecharim Works
Descriptiondescription / how-mecharim-works

How Mecharim Works

Learn how Mecharim turns a business into a digital twin and lets Mechas operate inside that grounded world.

Read this after the system overview if you want the clearest explanation of the business twin, the Mecha layer, and the control-plane versus runtime split.

Before you continue

Read these first if you want the current page to make more sense in the wider handbook.

Two halves of one world

The shortest correct explanation of Mecharim is this:

First, the platform models a real business as a structured world. Second, it gives that world named Mechas that can communicate, retrieve grounded context, and do work inside it.

That is why the system has two major sides:

  • the business twin
  • the Mecha layer

Part 1. The business twin

The business twin is the structured representation of a real business:

text
Organization -> Base -> Unit -> Anchor -> Xenkey

This chain is the backbone that tells the platform:

  • who the business is
  • where it operates
  • how it is organized
  • what real things belong to it
  • what those things mean

Why this matters

Without this structure, business reality reaches AI mostly as:

  • scattered pages
  • disconnected descriptions
  • mixed ownership
  • weakly structured metadata

With this structure, the system gains a clearer and more durable world model.

Part 2. The Mecha layer

The Mecha layer is the acting layer:

text
Crew -> Mecha

This layer gives the system named AI workers that can:

  • communicate
  • answer
  • retrieve grounded context
  • support workflows
  • coordinate with humans and other Mechas

These workers do not replace the business twin. They depend on it.

The most important distinction

The business twin and the Mecha layer are not competing models. They are two halves of one operational system.

The right sentence to remember is:

The digital twin describes the world. Mechas work inside the world that has been described.

Why Anchor and Xenkey matter so much

Many systems can store text. Much fewer systems can bind meaning to owned business reality cleanly.

That is the function of the Anchor and Xenkey pair.

Anchor

An Anchor fixes something real into the model, for example:

  • a product
  • a service
  • a person
  • a team
  • a process
  • a location

The point of Anchor is simple:

Anchor binds reality into the platform.

Xenkey

A Xenkey is the semantic layer attached to that anchored thing.

It gives the system structured meaning that is not floating on its own.

The correct relationship is:

Anchor binds reality. Xenkey binds meaning to that reality.

Why the pair matters

Without Anchor + Xenkey:

  • knowledge becomes generic
  • search becomes fuzzy
  • support drifts
  • AI actions become harder to explain

With Anchor + Xenkey:

  • meaning stays attached to real entities
  • retrieval becomes more precise
  • support becomes more contextual
  • AI behavior stays closer to business ownership

Operator flow

From the operator point of view, the system works like this:

  1. Create or manage the business structure.
  2. Define the digital twin through Organization, Base, Unit, Anchor, and Xenkey.
  3. Create the Mecha layer through Crew and Mecha.
  4. Issue credentials to the Mecha.
  5. Connect an external runtime.
  6. Let the Mecha use communication, retrieval, and workflow paths.

This is why the frontend acts as the control plane.

Runtime flow

From the runtime point of view, the system works like this:

  1. A Mecha exists under a Crew.
  2. A key is issued to that Mecha.
  3. The external runtime receives the credential.
  4. The runtime connects over REST and WebSocket.
  5. It communicates through Mechagram.
  6. It may retrieve grounded knowledge through MechaHub.
  7. It acts inside the structured business world.

This is why the runtime side is much more than “chat”. It is identity, transport, retrieval, grounding, and operational behavior.

How the layers interact

The full interaction looks like this:

  1. The business is modeled as structured data.
  2. Real entities are fixed through Anchors.
  3. Meaning is attached through Xenkeys.
  4. Knowledge and retrieval use that grounded structure.
  5. Mechas operate in that environment through identity and transport.
  6. Discovery can project selected public reality through MechaReg.

The system becomes useful because these steps reinforce each other instead of competing for ownership.

Business effect

The combined system gives something stronger than either half alone:

  • not just stored data
  • not just AI workers
  • but AI workers operating inside a grounded and structured business world

That is what makes the platform operationally useful.