Create Accountor Login
Docs

The Permission Catalog

Part of the MountDev Ops Workspace documentation. Updated November 17, 2026

MountDev Ops Workspace documentation | Updated November 17, 2026

Our Commitment MountDev Ops is built security first. Your business data belongs to you, and you can export all of it at any time. We never use analytics or tracking of any kind inside MountDev Ops.

The permission catalog is the one list of what each app can be given: its kinds of records, the levels each kind has, and which actions are which. It is made from each app's own description of itself, so a new app appears in it with no permission screen of its own. It describes the product's actions only. It holds no company's data, and it never reads a company's database.

What It Describes

  • Kinds. A kind is a kind of record, such as Tickets in Helpdesk, or a named area for actions that are not about a record, such as Helpdesk Settings.
  • Levels. Each kind offers some of See, Add, Change, Remove, Export, Approve, and Spend. A level that nothing in an app does is not offered. A kind of record always has See, Add, Change, Remove, and Export.
  • Classes. A kind of record is customer, work, money, content, or setup. A sensitive field is prices, pay, bank, private, or logins.

How an App Declares It

An app says what it is in its manifest and nothing else:

  • a module declares its class;
  • an action declares its kind (a module id or an area such as helpdesk.settings) and, when it is not what its level says, its verb;
  • a field that holds prices, pay, bank details, or a private note declares its class;
  • an action only people may run declares peopleOnly: true, and is never marked for AI agents or API keys.

Each level maps to the connection level an action already has: See is read; Add, Change, Export, and Approve are write; Remove is delete; Spend is spending. The permission id and level every action declares for connections do not change.

Until an app declares them, an action's kind is its area and its verb follows its level. A list of those actions only shrinks, and a new action, module, or sensitive field must declare itself. Helpdesk and the core's records, roles, access, groups, people, and activity actions are declared today.

What It Changes Today

Nothing a person or an AI agent can do. The catalog is read by nothing that decides access yet. Who may do what is still decided by roles, permission groups, and the permission each action names, as described in Access and Permissions. The catalog is the first part of the permissions overhaul. Presets, scopes, "What can this person do?", and the Staff app's permission screens are built on it in later parts (Planned).

Ask Us Anything

We’d love to hear from you!