Skip to content
nara Documentation

Edge Connector

EDGE Desktop & Server CLI‑first

What the Edge Connector does

Section titled “What the Edge Connector does”

The nara Edge Connector bridges the nara platform with your own infrastructure:

  • runs your tool implementations next to the systems they talk to
  • maintains a WebSocket connection to the Tool RPC server
  • keeps deployments and credentials in sync with the nara webapp

Use it whenever a tool needs to reach on‑prem systems, private APIs, or local hardware.

High-level flow

Section titled “High-level flow”
  1. Authenticate the connector with your organization using edge-connector auth or the desktop bootstrap.
  2. Generate language-specific templates for tools you want to implement locally.
  3. Implement & test tools with edge-connector generate and edge-connector test.
  4. Package implementations into a signed bundle and upload it to the platform.
  5. Download preconfigured connector bundles from the webapp and deploy them on target machines.

Key building blocks

Section titled “Key building blocks”

CLI

The edge-connector CLI handles authentication, code generation, testing, packaging, and uploads. See the CLI page for details.

Runtime

A long‑running Node.js process that connects to Tool RPC, loads your implementations, and executes tools on demand.

Deployments

Each running connector instance registers as a deployment in the webapp so you can see status, last seen time, and configuration.

Bundles

Uploadable .tgz archives containing your compiled implementations and a manifest, created by the edge-connector package command.

Downloads

Platform-generated archives (per OS) that bundle the runtime, your latest tools, and a provisioning.json file with deployment settings.

Auto-Resume

The connector persists its last known configuration and can restart with --auto-resume to reuse the previous Tool RPC settings and deployment.

Where to go next

Section titled “Where to go next”

Start with the Edge Connector CLI