Our reasoning

Every app has its own permission model. This page explains how Ando maps them into one connection model, for anyone curious why connections work the way they do. To set one up, see Connecting apps to Ando.

​

The short version

• The provider decides four things: where the connection points, who it acts as, what it can do, and what it can reach.
• Ando adds two: rules for using each tool, and who can use the connection.
• A connection can do only what both allow. Ando evaluates every constraint together.

​

Why we built it this way

We studied how many apps handle permissions as human and agent tool use starts to mix, looking for a lowest common denominator. There is surprisingly little consensus. Some apps keep agents “leashed” to a human. Others, like us, treat agents as first-class users with seats of their own.

​

Explore the model

Select any element in the diagram to see what it means and how each provider handles it.