Prime Intellect · Capability

Prime Intellect Evaluations API — Evaluations

Prime Intellect Evaluations API — Evaluations. 7 operations. Self-contained Naftiko capability covering one Prime Intellect business surface.

Prime Intellect Evaluations API — Evaluations is a Naftiko capability published by Prime Intellect, one of 13 capabilities the APIs.io network indexes for this provider. It bundles 7 operations across the POST, GET, and DELETE methods rooted at /v1/api/v1/evaluations.

The capability includes 3 read-only operations and 4 state-changing operations. Lead operation: Create a new evaluation. Can be deployed as a REST endpoint, MCP tool, or Agent Skill via Naftiko.

Tagged areas include Prime Intellect and Evaluations.

Run with Naftiko Prime IntellectEvaluations

What You Can Do

POST
Createeval — Create a new evaluation.
/v1/api/v1/evaluations/
GET
Listevals — List evaluations.
/v1/api/v1/evaluations/
GET
Geteval — Get a single evaluation.
/v1/api/v1/evaluations/{evaluation_id}
DELETE
Deleteeval — Delete an evaluation.
/v1/api/v1/evaluations/{evaluation_id}
POST
Finalizeeval — Finalize an evaluation.
/v1/api/v1/evaluations/{evaluation_id}/finalize
POST
Pushsamples — Push samples to an evaluation.
/v1/api/v1/evaluations/{evaluation_id}/samples
GET
Getsamples — Retrieve evaluation samples.
/v1/api/v1/evaluations/{evaluation_id}/samples

MCP Tools

prime-intellect-createeval

Create a new evaluation.

prime-intellect-listevals

List evaluations.

read-only idempotent
prime-intellect-geteval

Get a single evaluation.

read-only idempotent
prime-intellect-deleteeval

Delete an evaluation.

idempotent
prime-intellect-finalizeeval

Finalize an evaluation.

prime-intellect-pushsamples

Push samples to an evaluation.

prime-intellect-getsamples

Retrieve evaluation samples.

read-only idempotent

Capability Spec

evals-evaluations.yaml Raw ↑ 
naftiko: 1.0.0-alpha2
info:
  label: Prime Intellect Evaluations API — Evaluations
  description: Prime Intellect Evaluations API — Evaluations. 7 operations. Self-contained Naftiko capability covering one
    Prime Intellect business surface.
  tags:
  - Prime Intellect
  - Evaluations
  created: '2026-05-25'
  modified: '2026-05-25'
binds:
- namespace: env
  keys:
    PRIME_API_KEY: PRIME_API_KEY
capability:
  consumes:
  - type: http
    namespace: evals-evaluations
    baseUri: https://api.primeintellect.ai
    description: Prime Intellect Evaluations API — Evaluations business capability. Self-contained, no shared references.
    resources:
    - name: evaluations
      path: /api/v1/evaluations/
      operations:
      - name: createeval
        method: POST
        description: Create a new evaluation.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: body
          in: body
          type: object
          description: Request body (JSON).
          required: true
      - name: listevals
        method: GET
        description: List evaluations.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
    - name: evaluations-evaluation_id
      path: /api/v1/evaluations/{evaluation_id}
      operations:
      - name: geteval
        method: GET
        description: Get a single evaluation.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: evaluation_id
          in: path
          type: string
          required: true
      - name: deleteeval
        method: DELETE
        description: Delete an evaluation.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: evaluation_id
          in: path
          type: string
          required: true
    - name: evaluations-evaluation_id-finalize
      path: /api/v1/evaluations/{evaluation_id}/finalize
      operations:
      - name: finalizeeval
        method: POST
        description: Finalize an evaluation.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: body
          in: body
          type: object
          description: Request body (JSON).
          required: true
        - name: evaluation_id
          in: path
          type: string
          required: true
    - name: evaluations-evaluation_id-samples
      path: /api/v1/evaluations/{evaluation_id}/samples
      operations:
      - name: pushsamples
        method: POST
        description: Push samples to an evaluation.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: body
          in: body
          type: object
          description: Request body (JSON).
          required: true
        - name: evaluation_id
          in: path
          type: string
          required: true
      - name: getsamples
        method: GET
        description: Retrieve evaluation samples.
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: evaluation_id
          in: path
          type: string
          required: true
    authentication:
      type: bearer
      value: '{{env.PRIME_API_KEY}}'
      placement: header
  exposes:
  - type: rest
    namespace: evals-evaluations-rest
    port: 8080
    description: REST adapter for Prime Intellect Evaluations API — Evaluations.
    resources:
    - path: /v1/api/v1/evaluations/
      name: evaluations
      description: REST surface for evaluations.
      operations:
      - method: POST
        name: createeval
        description: Create a new evaluation.
        call: evals-evaluations.createeval
        with:
          body: rest.body
        outputParameters:
        - type: object
          mapping: $.
      - method: GET
        name: listevals
        description: List evaluations.
        call: evals-evaluations.listevals
        with: {}
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/api/v1/evaluations/{evaluation_id}
      name: evaluations-evaluation_id
      description: REST surface for evaluations-evaluation_id.
      operations:
      - method: GET
        name: geteval
        description: Get a single evaluation.
        call: evals-evaluations.geteval
        with:
          evaluation_id: rest.path.evaluation_id
        outputParameters:
        - type: object
          mapping: $.
      - method: DELETE
        name: deleteeval
        description: Delete an evaluation.
        call: evals-evaluations.deleteeval
        with:
          evaluation_id: rest.path.evaluation_id
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/api/v1/evaluations/{evaluation_id}/finalize
      name: evaluations-evaluation_id-finalize
      description: REST surface for evaluations-evaluation_id-finalize.
      operations:
      - method: POST
        name: finalizeeval
        description: Finalize an evaluation.
        call: evals-evaluations.finalizeeval
        with:
          body: rest.body
          evaluation_id: rest.path.evaluation_id
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/api/v1/evaluations/{evaluation_id}/samples
      name: evaluations-evaluation_id-samples
      description: REST surface for evaluations-evaluation_id-samples.
      operations:
      - method: POST
        name: pushsamples
        description: Push samples to an evaluation.
        call: evals-evaluations.pushsamples
        with:
          body: rest.body
          evaluation_id: rest.path.evaluation_id
        outputParameters:
        - type: object
          mapping: $.
      - method: GET
        name: getsamples
        description: Retrieve evaluation samples.
        call: evals-evaluations.getsamples
        with:
          evaluation_id: rest.path.evaluation_id
        outputParameters:
        - type: object
          mapping: $.
  - type: mcp
    namespace: evals-evaluations-mcp
    port: 9090
    transport: http
    description: MCP adapter for Prime Intellect Evaluations API — Evaluations. One tool per consumed operation, routed inline
      through this capability's consumes block.
    tools:
    - name: prime-intellect-createeval
      description: Create a new evaluation.
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: evals-evaluations.createeval
      with:
        body: tools.body
      outputParameters:
      - type: object
        mapping: $.
    - name: prime-intellect-listevals
      description: List evaluations.
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: evals-evaluations.listevals
      with: {}
      outputParameters:
      - type: object
        mapping: $.
    - name: prime-intellect-geteval
      description: Get a single evaluation.
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: evals-evaluations.geteval
      with:
        evaluation_id: tools.evaluation_id
      outputParameters:
      - type: object
        mapping: $.
    - name: prime-intellect-deleteeval
      description: Delete an evaluation.
      hints:
        readOnly: false
        destructive: true
        idempotent: true
      call: evals-evaluations.deleteeval
      with:
        evaluation_id: tools.evaluation_id
      outputParameters:
      - type: object
        mapping: $.
    - name: prime-intellect-finalizeeval
      description: Finalize an evaluation.
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: evals-evaluations.finalizeeval
      with:
        body: tools.body
        evaluation_id: tools.evaluation_id
      outputParameters:
      - type: object
        mapping: $.
    - name: prime-intellect-pushsamples
      description: Push samples to an evaluation.
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: evals-evaluations.pushsamples
      with:
        body: tools.body
        evaluation_id: tools.evaluation_id
      outputParameters:
      - type: object
        mapping: $.
    - name: prime-intellect-getsamples
      description: Retrieve evaluation samples.
      hints:
        readOnly: true
        destructive: false
        idempotent: true
      call: evals-evaluations.getsamples
      with:
        evaluation_id: tools.evaluation_id
      outputParameters:
      - type: object
        mapping: $.