> ## Documentation Index
> Fetch the complete documentation index at: https://docs.open-cluster.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Connect your tools

> Add read-only integrations and verify what evidence OpenCluster can retrieve.

An Integration is one configured installation in an Organization. Choose at least one
alert source to create Incidents and one evidence source that can help explain them.
OpenCluster requests read-only access and never changes a connected system.

| What you need                         | Integration             | Evidence or intake                                                | Relay required | Setup guide                                                    |
| ------------------------------------- | ----------------------- | ----------------------------------------------------------------- | -------------- | -------------------------------------------------------------- |
| Send canonical alerts from any system | Generic Webhook         | Firing and resolved Alert Events                                  | No             | [Generic Webhook](/integrations/alerting/generic_webhook)      |
| Receive Prometheus alert lifecycles   | Prometheus Alertmanager | Grouped firing and resolved alerts                                | No             | [Prometheus Alertmanager](/integrations/alerting/alertmanager) |
| Inspect cluster runtime state         | Kubernetes              | Workloads, Events, and bounded pod logs                           | Yes            | [Kubernetes](/integrations/infrastructure/kubernetes)          |
| Compare incidents with code changes   | GitHub                  | Repositories, commits, pull requests, checks, files, and releases | No             | [GitHub](/integrations/source-control/github)                  |
| Read incident conversation context    | Slack                   | Bounded channel and thread messages                               | No             | [Slack](/integrations/collaboration/slack)                     |

## Prerequisites

You need an **Admin** membership in the active Organization and the provider permissions
listed in the selected setup guide. Kubernetes also requires a deployed Relay with an
outbound connection to the control plane.

## Connect and verify

1. Open **Integrations** and choose the Integration.
2. Enter the required non-secret configuration and credential, or complete the provider
   authorization flow when the deployment offers one.
3. Select **Verify** and review the available evidence and any limitations.
4. For an inbound alert Integration, send one test delivery.

Credentials are write-only after submission. Webhook secrets are displayed once. A
successful read Integration becomes `active`; it can be `degraded` when only some
evidence is available. A webhook Integration remains `configured` until its first valid
delivery and then becomes `active`.

If verification fails, check the provider URL, credential, least-privilege permissions,
and network access before replacing the credential. Verification proves current access;
it does not guarantee that a later Investigation will find enough evidence for a cause.

## Disable, rotate, or remove

Disable an Integration to exclude it from new Investigations while retaining history.
Replace its credential through the provider guide's rotation procedure. Deletion is
refused while retained records depend on the Integration.

Next, [run your first investigation](/getting-started/run-your-first-investigation).