Getting Started

This is the 15-minute path: Compose up, admin account, a service that can page you, one test incident. Use the other guides in this section when you need install variants or a fuller first-week setup.

Minutes	What you do	Done when
0–5	Run Compose and create the admin	You can sign in at `http://localhost:3000`
5–12	Team → schedule → policy → service	The service routes to your current schedule
12–15	Open a test incident	It appears OPEN; you can acknowledge it

Notifications (email/SMS/Slack) are optional for this path. Without a provider, the incident still exists; nobody is paged off-box.

1. Run OpsKnight (Compose)

bash

git clone https://github.com/opsknight-labs/OpsKnight.git
cd OpsKnight
cp env.example .env

Generate the two secrets, then paste the output values into .env before you start:

bash

openssl rand -base64 32
openssl rand -hex 32

dotenv

NEXTAUTH_URL=http://localhost:3000
NEXT_PUBLIC_APP_URL=http://localhost:3000
NEXTAUTH_SECRET=PASTE_BASE64_OUTPUT
ENCRYPTION_KEY=PASTE_64_HEX_CHARACTER_OUTPUT

Dotenv files do not evaluate $(...) shell substitutions.

The Compose file constructs the application's container-only DATABASE_URL from POSTGRES_USER, POSTGRES_PASSWORD, and POSTGRES_DB; you do not need to change the host-development DATABASE_URL example for this path.

bash

docker compose up -d
open http://localhost:3000

First boot sends you to /setup. Create the admin, copy the generated password once, then sign in.

Helm, Kustomize, and from-source installs: Installation. Env reference: Configuration.

2. Make something that can page

Stay on this machine. You do not need a second user for a first page.

Teams — create Platform and add yourself.
Schedules — create Platform primary, add yourself on a 24×7 layer (or a short override covering now).
Policies — create Platform page, one step: notify the schedule, a few minutes before the next step (you will ack before that).
Services — create Checkout API, attach that policy.

Detail and screenshots: First Steps.

3. Fire a test incident

In the UI

Incidents → Create Incident
Title Test page, service Checkout API, urgency High
Create

You should see status OPEN and a timeline entry. Click Acknowledge, then Resolve.

Optional: Events API (after you add an integration key on the service)

bash

curl -X POST http://localhost:3000/api/events \
  -H "Content-Type: application/json" \
  -H "Authorization: Token token=YOUR_INTEGRATION_KEY" \
  -d '{
    "event_action": "trigger",
    "dedup_key": "getting-started-test-1",
    "payload": {
      "summary": "Test page from getting started",
      "severity": "critical",
      "source": "getting-started"
    }
  }'

See Events API.

4. Confirm the install

bash

curl -s http://localhost:3000/api/health

Expect "status":"healthy" (or "degraded" only if a non-database check failed). If the UI never loads, Troubleshooting.

After the first page

Next job	Guide
Email / SMS / Slack so a real page leaves the box	Notifications
One public/private status page	Status page
Slack war rooms (this version)	Slack ChatOps
Ingest from monitoring	Integrations
OIDC SSO (not SAML)	OIDC

There is no voice channel. Microsoft Teams and Google Chat are webhook formats, not Slack-style rooms.