Major is built around several core concepts that work together to help you build, deploy, and secure internal tools.

​

Apps

An App is a self-contained project that you build and deploy through Major.

​

Lifecycle

Apps follow a simple lifecycle:

1. Create — Scaffold from a template or clone an existing app
2. Develop — Build locally with the CLI or in the web editor
3. Connect — Attach connectors (databases, APIs)
4. Deploy — Ship to production with one command or click
5. Share — Grant access to teammates via RBAC

​

Source Control

Every app is backed by a Git repository. Major hosts repositories by default, but you can connect your own GitHub organization at any time. Either way, the code is yours — clone, branch, and manage it with any Git tooling.

​

Connectors

Connectors are the databases, APIs, and storage services your apps connect to. Major manages credentials securely so you never need to handle secrets in your code.

​

Architecture

• Secure Flow: Your app connects to the Major API, which proxies requests to your actual connectors.
• Authentication: Powered by a MAJOR_JWT_TOKEN issued to your session. This token contains your user scopes, ensuring the client can only access what you’re permitted to access.
• Generated Clients: Major generates resource clients for you. You’ll find one per resource in src/clients/.

​

RBAC (Role-Based Access Control)

Major provides granular access control at both the organization and application level.

​

Organization roles

​

Application roles

​

Resource roles

​

Groups

Groups let you manage permissions for sets of users efficiently:

• Inheritance — Group members inherit all permissions granted to the group
• Default permissions — Configure groups to have automatic access to new apps
• All Users group — Every organization member, useful for open-by-default apps

​

Memory

Persistent memory stores organization-wide knowledge that your AI assistants and agents can access and update across sessions. Use memory to capture business context, resource quirks, and preferences without re-explaining them each time. Memory is organized into files that can be viewed, created, edited, and deleted — either through the Memory panel in the editor or through API endpoints. The system tracks which source (user, AI coder, or memory agent) made each change.

​

How they work together

1. Admins configure connectors and grant permissions to users and groups
2. Builders create applications and use connectors to securely access company data
3. Users get access to apps via RBAC and use them to get work done

​

Next steps