Skip to main content

Contagious Layers

The Power of Primitives​

In Synap, extensibility isn't an afterthought—it's contagious.

When you add a new primitive to the system (like a new Entity Type, or a new Agent), you don't just add a standalone feature. You trigger a cascade of capabilities that "infects" every layer of the operating system.

How It Works​

Let's say you define a new Entity Type: Book.

1. The Infection Begins (The Definition)​

You simply define the schema:

{
  type: 'book',
  fields: { title, author, isbn, rating }
}

2. The Spread (Automatic Capabilities)​

Instantly, without writing extra code, your Book entity gains:

  • 🧠 The Intelligence Layer

    • The Hub Protocol exposes Book to all AI agents.
    • Agents can now read, create, and update books.
    • They understand the schema automatically via introspection.

  • 🕸️ The Graph Layer

    • Book nodes appear in the Knowledge Graph.
    • You can link a Task to a Book.
    • You can see all Notes referencing a Book.

  • 🔍 The Search Layer

    • Vector embeddings are generated for Book content.
    • Semantic search ("novels about space") finds your books.
    • Full-text search indices are updated.

  • ⚡ The Realtime Layer

    • WebSocket channels (entities.book.*) are created.
    • Collaborators see updates instantly.

  • 🛡️ The Security Layer

    • Standard permissions (Owner, Editor, Viewer) apply immediately.
    • Global Validator protects Book creation policies.

"Write Once, Propagate Everywhere"​

This architectural property is what makes Synap a true OS.

  • Legacy App: Add a feature → Update Database → Update API → Update Frontend → Update Search → Update Mobile App.
  • Synap: Add a Primitive → The System Adapts.

This creates a flywheel effect: The more primitives you add, the richer the entire ecosystem becomes for every other primitive. Your Book entity makes the Research Agent smarter, which makes your Notes more valuable, which makes the Graph more insightful.