Services

PRISM models services as first-class entities that link teams, layers, and metrics together. Services represent deployable units that teams own and operate.

Service Definition

Field	Description
`id`	Unique identifier
`name`	Display name
`description`	Service description
`ownerTeamId`	Team responsible for this service
`layerId`	Primary layer (code, infra, runtime)
`metricIds`	Metrics associated with this service
`repository`	Git repository URL
`tier`	Service tier (tier1, tier2, tier3)

Basic Example

{
  "services": [
    {
      "id": "payments-api",
      "name": "Payments API",
      "description": "Core payments processing service",
      "ownerTeamId": "payments-team",
      "layerId": "runtime",
      "tier": "tier1",
      "repository": "https://github.com/example/payments-api"
    }
  ]
}

Service Tiers

Service tiers indicate criticality and SLO expectations:

Tier	Description	Typical SLO
tier1	Business-critical services	99.99%
tier2	Important services	99.9%
tier3	Internal/supporting services	99%

Linking Services to Metrics

Services can be linked to metrics in two ways:

1. Service MetricIDs Array

{
  "id": "payments-api",
  "metricIds": ["payments-availability", "payments-latency"]
}

2. Metric ServiceID Field

{
  "id": "payments-availability",
  "name": "Payments Availability",
  "serviceId": "payments-api",
  "domain": "operations",
  "stage": "runtime"
}

Service Ownership

Services connect to teams through the ownerTeamId field:

{
  "teams": [
    {
      "id": "payments-team",
      "name": "Payments Team",
      "type": "stream_aligned",
      "serviceIds": ["payments-api", "payments-worker"]
    }
  ],
  "services": [
    {
      "id": "payments-api",
      "ownerTeamId": "payments-team"
    }
  ]
}

Full Example

{
  "teams": [
    {
      "id": "payments-team",
      "name": "Payments Team",
      "type": "stream_aligned"
    }
  ],
  "services": [
    {
      "id": "payments-api",
      "name": "Payments API",
      "description": "Handles payment processing",
      "ownerTeamId": "payments-team",
      "layerId": "runtime",
      "tier": "tier1",
      "metricIds": ["slo-payments-availability"]
    }
  ],
  "metrics": [
    {
      "id": "slo-payments-availability",
      "name": "Payments Availability",
      "domain": "operations",
      "stage": "runtime",
      "category": "reliability",
      "layer": "runtime",
      "serviceId": "payments-api",
      "metricType": "rate",
      "current": 99.95,
      "target": 99.99,
      "slo": {
        "target": ">=99.99%",
        "operator": "gte",
        "value": 99.99,
        "window": "30d"
      }
    }
  ]
}

CLI Commands

List services in a document:

prism service list prism.json

Show service details with metrics:

prism service show prism.json payments-api

Best Practices

Define all production services - Complete service catalog
Assign owner teams - Clear accountability
Set service tiers - Differentiate criticality
Link to metrics - Connect services to their SLOs
Include repository URLs - Enable code navigation
Use consistent naming - Service IDs should match deployment names