Skip to content

📈 Enterprise Observability: Monitoring & Metrics ​

Artifact-Shield is built for mission-critical infrastructure. To ensure your gateway is healthy and performing at scale, use the built-in observability features.


🚀 1. Prometheus & Grafana Integration ​

The gateway exposes a standard Prometheus endpoint containing real-time metrics for garbage collection, memory usage, and custom application metrics.

Endpoint ​

GET /actuator/prometheus

Setup Guide ​

  1. Prometheus Config: Add Artifact-Shield to your prometheus.yml:
    yaml
    scrape_configs:
      - job_name: 'artifact-shield'
        metrics_path: '/actuator/prometheus'
        static_configs:
          - targets: ['shield-gateway:8080']
  2. Grafana Dashboard: Import a standard "Spring Boot 3" dashboard (e.g., ID 19022) to visualize:
    • JVM Heap Usage
    • HTTP Response Times (p95, p99)
    • Thread Pool Saturation

🛡️ 2. Custom Business Metrics ​

Artifact-Shield exposes several custom metrics to track the security of your AI traffic:

Metric NameDescription
shield.sanitize.countTotal number of scans performed.
shield.risk.avgRolling average of the risk scores detected.
shield.redactions.totalCumulative count of PII strings redacted.
shield.proxy.countNumber of successful downstream LLM calls.

These are viewable via GET /actuator/metrics.


🔎 3. Splunk Integration ​

For security forensics, enable the built-in Splunk HEC (HTTP Event Collector) logger.

Configuration ​

yaml
shield:
  logging:
    splunk:
      enabled: true
      url: "http://splunk-hec.corp:8088/services/collector/event"
      token: "YOUR-HEC-TOKEN"

Once enabled, every AuditLog generated by the gateway is asynchronously streamed to Splunk, allowing you to create Real-Time Security Alerting for critical PII leaks.


🩺 4. Health Checks & Liveness ​

Use the standard Actuator health endpoint for Kubernetes probes.

Endpoint: GET /actuator/health

Success Result (200 OK):

json
{
  "status": "UP",
  "components": {
    "db": { "status": "UP" },
    "diskSpace": { "status": "UP" },
    "ping": { "status": "UP" }
  }
}

For more help, contact your SRE team or visit the [Artifact-Shield internal wiki].