Skip to main content

​
Neurons

A neuron is a configurable agent: model selection, tools (reflexes), memory, and visibility scope (core, workspace, or brain).

​
Invoke via receptor

Production apps usually invoke neurons through receptors, not by calling neuron endpoints directly. Receptors handle triggers, routing, and share tokens.

​
Scope and visibility

ScopeMeaning
corePlatform-provided, read-only for most workspaces
workspaceShared across the workspace
brainScoped to a project (X-Brain-Id)
Details: Neurons API — scope.

​
SDK

const neurons = await auvy.neurons.list({ brain_id: brainId })
const neuron = await auvy.neurons.get(neuronId)