#  Copyright (c) 2026 Cisco Systems, Inc. and its affiliates
#  SPDX-License-Identifier: Apache-2.0
$schema: "http://json-schema.org/draft-07/schema#"
$id: "mas-lab/dataset/v1"
title: "mas-lab Dataset manifest (v1)"
description: >
  Dataset items use the Run Input Envelope (inputs + expectations).
  See run-input.schema.yaml for item shape.

type: object
required: [apiVersion, kind, metadata, spec]
additionalProperties: false

properties:
  apiVersion:
    type: string
    const: "lab/v1"

  kind:
    type: string
    const: "Dataset"

  metadata:
    type: object
    required: [name]
    additionalProperties: false
    properties:
      name:
        type: string
        minLength: 1
      display_name:
        type: string
      version:
        type: string
        default: "1.0"
      description:
        type: string
      tags:
        type: array
        items:
          type: string

  spec:
    type: object
    additionalProperties: false
    oneOf:
      - required: [items]
      - required: [path]
    properties:
      items:
        type: array
        minItems: 1
        items:
          type: object
          required: [id, inputs]
          additionalProperties: true
          properties:
            id:
              oneOf:
                - type: string
                - type: integer
            inputs:
              type: object
              required: [user]
              additionalProperties: true
              properties:
                user:
                  type: array
                  minItems: 1
                  items:
                    type: object
                    required: [role, content]
                    additionalProperties: false
                    properties:
                      role:
                        type: string
                        enum: [user, system]
                      content:
                        type: string
                        minLength: 1
                hitl:
                  type: array
                  items:
                    type: object
                    required: [role, content]
                    additionalProperties: false
                    properties:
                      role:
                        type: string
                        enum: [hitl]
                      content:
                        type: string
                memory_seeds:
                  oneOf:
                    - type: string
                    - type: array
                      items:
                        type: object
                        additionalProperties: true
                tool_fixtures:
                  oneOf:
                    - type: string
                    - type: object
                      additionalProperties: true
                session_id:
                  type: string
                checkpoint:
                  type: object
                  additionalProperties: true
            expectations:
              type: object
              additionalProperties: true
              properties:
                ground_truth:
                  oneOf:
                    - type: string
                    - type: object
                      additionalProperties: true
                governance:
                  type: object
                  additionalProperties: true
                metrics:
                  type: array
                  items:
                    type: string
            category:
              type: string
            group:
              type: string
            type:
              type: string
            tags:
              type: array
              items:
                type: string
            metadata:
              type: object
              additionalProperties: true

      path:
        type: string
      format:
        type: string
        enum: [yaml]
        default: yaml
      item_schema:
        type: object
        additionalProperties: true
      groups:
        type: array
        items:
          type: object
          required: [id]
          additionalProperties: false
          properties:
            id:
              type: string
            description:
              type: string
            count:
              type: integer