See It In Action
Watch NthLayer transform a service definition into complete observability
Define Your Service
Simple YAML file describing your service and dependencies:
service:
name: payment-api
tier: critical
description: Payment processing service
dependencies:
databases:
- type: postgresql
name: payments-db
- type: redis
name: session-cache
external:
- name: stripe-api
critical: trueRun One Command
Generate everything with a single unified command:
$ nthlayer apply payment-api.yaml
🚀 Analyzing service: payment-api
✅ [1/5] SLOs → 3 created
• availability (99.9%)
• latency-p95 (500ms)
• latency-p99 (1000ms)
✅ [2/5] Alerts → 28 created
• 8 SLO burn rate alerts
• 6 service health alerts
• 14 dependency alerts
✅ [3/5] Dashboard → 1 created
• 12 panels (health, SLOs, dependencies)
• Auto-configured for Grafana
✅ [4/5] Recording → 21 created
• 10x faster dashboard queries
✅ [5/5] PagerDuty → configured
• Team, schedules, escalation policy
• Tier-based timing (5/15/30min for critical)
✨ Done in 2.3sView Live Dashboard
Real dashboard running in Grafana Cloud, powered by live metrics from our demo app:
6 Live Services with Auto-Generated Dashboards
Each dashboard is organized into: SLO Metrics → Service Health → Dependencies
Click any service to view its auto-generated Grafana dashboard
🚨 Auto-Generated Prometheus Alerts
118 production-ready alerts across 6 services, sourced from awesome-prometheus-alerts
Click any service to view its generated Prometheus alert rules on GitHub
🚀 Deploy Your Alerts
Generated alerts are standard Prometheus format — deploy anywhere
Using Prometheus Operator? Apply as a PrometheusRule CRD:
# Wrap in PrometheusRule CRD
cat <<EOF | kubectl apply -f -
apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
name: payment-api-alerts
labels:
release: prometheus
spec:
$(cat generated/payment-api/alerts.yaml | sed 's/^/ /')
EOFPush directly to Grafana Mimir or Cortex ruler:
# Using mimirtool
mimirtool rules load \
--address=https://mimir.example.com \
--id=your-tenant \
generated/payment-api/alerts.yaml
# Or via API
curl -X POST https://mimir.example.com/prometheus/config/v1/rules/payment-api \
-H "X-Scope-OrgID: your-tenant" \
--data-binary @generated/payment-api/alerts.yamlCommit to your GitOps repo — ArgoCD or Flux handles the rest:
# Add to your GitOps repository
cp generated/payment-api/alerts.yaml \
k8s/monitoring/rules/payment-api.yaml
git add k8s/monitoring/rules/
git commit -m "Add payment-api alerting rules"
git push
# ArgoCD/Flux automatically syncs to cluster
# No manual kubectl required!Traditional Prometheus? Copy to rules directory and reload:
# Copy to Prometheus rules directory
cp generated/payment-api/alerts.yaml \
/etc/prometheus/rules/payment-api.yaml
# Validate syntax
promtool check rules /etc/prometheus/rules/*.yaml
# Reload Prometheus config
curl -X POST http://localhost:9090/-/reload📟 PagerDuty Integration
Complete incident response setup with tier-based escalation policies
Team Management
Auto-creates teams with manager roles assigned to API key owner
On-Call Schedules
Primary, secondary, and manager schedules with weekly rotation
Tier-Based Timing
Critical: 5→15→30min | High: 15→30→60min | Low: 60min only
Service Linking
Services linked to escalation policies with urgency settings
Team with manager role
Tier-based escalation (5/15/30min)
Service linked to policy
Support Models
- Everything generated from the same source (guaranteed consistency)
- Alerts automatically match SLO thresholds
- Dashboard panels linked to relevant alerts
- Recording rules optimize dashboard performance
- One command updates everything when service changes