Triton Inference Server · Capability

Triton Inference Server NVIDIA Triton Inference Server Metrics API — Metrics

Triton Inference Server NVIDIA Triton Inference Server Metrics API — Metrics. 1 operations. Lead operation: Triton Inference Server Get Prometheus metrics. Self-contained Naftiko capability covering one Triton business surface.

Run with Naftiko TritonMetrics

What You Can Do

GET
Getmetrics — Triton Inference Server Get Prometheus metrics
/v1/metrics

MCP Tools

triton-inference-server-get-prometheus

Triton Inference Server Get Prometheus metrics

read-only idempotent

Capability Spec

metrics-metrics.yaml Raw ↑
naftiko: 1.0.0-alpha2
info:
  label: Triton Inference Server NVIDIA Triton Inference Server Metrics API — Metrics
  description: 'Triton Inference Server NVIDIA Triton Inference Server Metrics API — Metrics. 1 operations. Lead operation:
    Triton Inference Server Get Prometheus metrics. Self-contained Naftiko capability covering one Triton business surface.'
  tags:
  - Triton
  - Metrics
  created: '2026-05-19'
  modified: '2026-05-19'
binds:
- namespace: env
  keys:
    TRITON_API_KEY: TRITON_API_KEY
capability:
  consumes:
  - type: http
    namespace: metrics-metrics
    baseUri: http://localhost:8002
    description: Triton Inference Server NVIDIA Triton Inference Server Metrics API — Metrics business capability. Self-contained,
      no shared references.
    resources:
    - name: metrics
      path: /metrics
      operations:
      - name: getmetrics
        method: GET
        description: Triton Inference Server Get Prometheus metrics
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
  exposes:
  - type: rest
    namespace: metrics-metrics-rest
    port: 8080
    description: REST adapter for Triton Inference Server NVIDIA Triton Inference Server Metrics API — Metrics. One Spectral-compliant
      resource per consumed operation, prefixed with /v1.
    resources:
    - path: /v1/metrics
      name: metrics
      description: REST surface for metrics.
      operations:
      - method: GET
        name: getmetrics
        description: Triton Inference Server Get Prometheus metrics
        call: metrics-metrics.getmetrics
        outputParameters:
        - type: object
          mapping: $.
  - type: mcp
    namespace: metrics-metrics-mcp
    port: 9090
    transport: http
    description: MCP adapter for Triton Inference Server NVIDIA Triton Inference Server Metrics API — Metrics. One tool per
      consumed operation, routed inline through this capability's consumes block.
    tools:
    - name: triton-inference-server-get-prometheus
      description: Triton Inference Server Get Prometheus metrics
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: metrics-metrics.getmetrics
      outputParameters:
      - type: object
        mapping: $.