Triton Inference Server · Capability

Triton Inference Server NVIDIA Triton Inference Server HTTP/REST API — Health

Triton Inference Server NVIDIA Triton Inference Server HTTP/REST API — Health. 4 operations. Lead operation: Triton Inference Server Server Liveness Check. Self-contained Naftiko capability covering one Triton business surface.

Run with Naftiko TritonHealth

What You Can Do

GET
Serverlive — Triton Inference Server Server Liveness Check
/v1/v2/health/live
GET
Serverready — Triton Inference Server Server Readiness Check
/v1/v2/health/ready
GET
Modelready — Triton Inference Server Model Readiness Check
/v1/v2/models/{model-name}/ready
GET
Modelversionready — Triton Inference Server Model Version Readiness Check
/v1/v2/models/{model-name}/versions/{model-version}/ready

MCP Tools

triton-inference-server-server-liveness

Triton Inference Server Server Liveness Check

read-only idempotent
triton-inference-server-server-readiness

Triton Inference Server Server Readiness Check

read-only idempotent
triton-inference-server-model-readiness

Triton Inference Server Model Readiness Check

read-only idempotent
triton-inference-server-model-version

Triton Inference Server Model Version Readiness Check

read-only idempotent

Capability Spec

http-rest-health.yaml Raw ↑
naftiko: 1.0.0-alpha2
info:
  label: Triton Inference Server NVIDIA Triton Inference Server HTTP/REST API — Health
  description: 'Triton Inference Server NVIDIA Triton Inference Server HTTP/REST API — Health. 4 operations. Lead operation:
    Triton Inference Server Server Liveness Check. Self-contained Naftiko capability covering one Triton business surface.'
  tags:
  - Triton
  - Health
  created: '2026-05-19'
  modified: '2026-05-19'
binds:
- namespace: env
  keys:
    TRITON_API_KEY: TRITON_API_KEY
capability:
  consumes:
  - type: http
    namespace: http-rest-health
    baseUri: http://localhost:8000
    description: Triton Inference Server NVIDIA Triton Inference Server HTTP/REST API — Health business capability. Self-contained,
      no shared references.
    resources:
    - name: v2-health-live
      path: /v2/health/live
      operations:
      - name: serverlive
        method: GET
        description: Triton Inference Server Server Liveness Check
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
    - name: v2-health-ready
      path: /v2/health/ready
      operations:
      - name: serverready
        method: GET
        description: Triton Inference Server Server Readiness Check
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
    - name: v2-models-model_name-ready
      path: /v2/models/{model_name}/ready
      operations:
      - name: modelready
        method: GET
        description: Triton Inference Server Model Readiness Check
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
    - name: v2-models-model_name-versions-model_version-ready
      path: /v2/models/{model_name}/versions/{model_version}/ready
      operations:
      - name: modelversionready
        method: GET
        description: Triton Inference Server Model Version Readiness Check
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
  exposes:
  - type: rest
    namespace: http-rest-health-rest
    port: 8080
    description: REST adapter for Triton Inference Server NVIDIA Triton Inference Server HTTP/REST API — Health. One Spectral-compliant
      resource per consumed operation, prefixed with /v1.
    resources:
    - path: /v1/v2/health/live
      name: v2-health-live
      description: REST surface for v2-health-live.
      operations:
      - method: GET
        name: serverlive
        description: Triton Inference Server Server Liveness Check
        call: http-rest-health.serverlive
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/v2/health/ready
      name: v2-health-ready
      description: REST surface for v2-health-ready.
      operations:
      - method: GET
        name: serverready
        description: Triton Inference Server Server Readiness Check
        call: http-rest-health.serverready
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/v2/models/{model-name}/ready
      name: v2-models-model-name-ready
      description: REST surface for v2-models-model_name-ready.
      operations:
      - method: GET
        name: modelready
        description: Triton Inference Server Model Readiness Check
        call: http-rest-health.modelready
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/v2/models/{model-name}/versions/{model-version}/ready
      name: v2-models-model-name-versions-model-version-ready
      description: REST surface for v2-models-model_name-versions-model_version-ready.
      operations:
      - method: GET
        name: modelversionready
        description: Triton Inference Server Model Version Readiness Check
        call: http-rest-health.modelversionready
        outputParameters:
        - type: object
          mapping: $.
  - type: mcp
    namespace: http-rest-health-mcp
    port: 9090
    transport: http
    description: MCP adapter for Triton Inference Server NVIDIA Triton Inference Server HTTP/REST API — Health. One tool per
      consumed operation, routed inline through this capability's consumes block.
    tools:
    - name: triton-inference-server-server-liveness
      description: Triton Inference Server Server Liveness Check
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: http-rest-health.serverlive
      outputParameters:
      - type: object
        mapping: $.
    - name: triton-inference-server-server-readiness
      description: Triton Inference Server Server Readiness Check
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: http-rest-health.serverready
      outputParameters:
      - type: object
        mapping: $.
    - name: triton-inference-server-model-readiness
      description: Triton Inference Server Model Readiness Check
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: http-rest-health.modelready
      outputParameters:
      - type: object
        mapping: $.
    - name: triton-inference-server-model-version
      description: Triton Inference Server Model Version Readiness Check
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: http-rest-health.modelversionready
      outputParameters:
      - type: object
        mapping: $.