Kubernetes

Connecting a Kubernetes cluster to NeuBird lets investigations pull in cluster configuration — pods, deployments, services, and related objects — and run read-only kubectl-style lookups against it directly. Each connection addresses one cluster.

NeuBird reaches the cluster through a Teleport Machine ID bot rather than a direct API server endpoint, so the cluster never needs to be exposed publicly.

Prerequisites

You need a Teleport cluster with:

• A Teleport Machine ID bot whose role grants access to the target Kubernetes cluster.
• A bot join token for that bot.
• The target cluster registered in Teleport with a label NeuBird can select it by (see Step 2).

If you don’t already run Teleport in front of this cluster, see the Teleport setup guide first.

Step 1: Gather the Teleport Proxy URL and Bot Join Token

1. In Teleport, note your Teleport Proxy URL — the proxy server address, including port, without a scheme (e.g. teleport.example.com:443).
2. Create or reuse a Machine ID bot scoped to this cluster and generate a bot join token for it.

Step 2: Determine the App Label Selector

NeuBird selects the cluster by a Teleport label, not by cluster name. Provide either:

• Just the customer/cluster identifier, e.g. my-cluster, or
• The fully-qualified label, e.g. customer=my-cluster.

Both forms resolve to the same selector. Use whatever label your Teleport Kubernetes access rules key off of.

Connection Fields

Field	Required	Description
Name	Yes	A descriptive name for this connection
Description	No	An overview of what this connection is used for
Teleport Proxy URL	Yes	The Teleport proxy server URL, e.g. teleport.example.com:443
Bot Join Token	Yes	The Teleport Machine ID bot’s join token
App Label Selector	Yes	The Teleport label selector for the cluster, e.g. my-cluster or customer=my-cluster

---

For general instructions on creating and managing connections, see the Connections Overview.