{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "urn:orbiplex:schema:corpus-entry:v1",
  "title": "CorpusEntry v1",
  "description": "Machine-readable schema for curated corpus entries derived from accepted bundles or promoted knowledge artifacts.",
  "type": "object",
  "additionalProperties": true,
  "x-dia-workflow": "project",
  "x-dia-status": "draft",
  "x-dia-basis": [
    "doc/project/50-requirements/requirements-004.md",
    "doc/project/50-requirements/requirements-002.md",
    "doc/project/50-requirements/requirements-003.md"
  ],
  "required": [
    "schema/v",
    "entry/id",
    "source/type",
    "source/id",
    "content/ref",
    "domain/tags",
    "quality/grade",
    "risk/grade",
    "training/eligibility",
    "provenance/manifest"
  ],
  "properties": {
    "schema/v": {
      "const": 1,
      "description": "Schema version."
    },
    "entry/id": {
      "type": "string",
      "minLength": 1,
      "description": "Stable identifier of the curated corpus entry."
    },
    "source/type": {
      "type": "string",
      "enum": [
        "transcript-bundle",
        "knowledge-artifact",
        "archival-package"
      ],
      "description": "Primary source class from which the corpus entry was assembled."
    },
    "source/id": {
      "type": "string",
      "minLength": 1,
      "description": "Identifier of the primary source artifact."
    },
    "content/ref": {
      "type": "string",
      "minLength": 1,
      "description": "Stable reference to the curated content body."
    },
    "domain/tags": {
      "type": "array",
      "minItems": 1,
      "items": {
        "type": "string",
        "minLength": 1
      },
      "description": "Domain and topic tags assigned by curation."
    },
    "quality/grade": {
      "type": "string",
      "enum": [
        "low",
        "medium",
        "high"
      ],
      "description": "Curation quality assessment of the entry."
    },
    "risk/grade": {
      "type": "string",
      "enum": [
        "low",
        "moderate",
        "high"
      ],
      "description": "Risk classification relevant to later publication or training use."
    },
    "training/eligibility": {
      "type": "string",
      "enum": [
        "blocked",
        "needs-review",
        "approved"
      ],
      "description": "Training eligibility state assigned to the corpus entry."
    },
    "provenance/manifest": {
      "type": "string",
      "minLength": 1,
      "description": "Reference to provenance manifest sufficient to reconstruct source lineage."
    },
    "contains-human-origin": {
      "type": "boolean",
      "description": "Whether the curated entry preserves human-originated source material."
    },
    "language": {
      "type": "string",
      "minLength": 1,
      "description": "Primary language of the curated content."
    },
    "creator/refs": {
      "type": "array",
      "items": {
        "type": "string",
        "minLength": 1
      },
      "description": "Curator, secretary, or contributor references that should survive attribution-sensitive flows."
    },
    "policy_annotations": {
      "type": "object",
      "additionalProperties": true,
      "description": "Optional implementation-local annotations that do not change the core corpus-entry semantics."
    }
  },
  "allOf": [
    {
      "if": {
        "properties": {
          "risk/grade": {
            "const": "high"
          }
        },
        "required": [
          "risk/grade"
        ]
      },
      "then": {
        "properties": {
          "training/eligibility": {
            "enum": [
              "blocked",
              "needs-review"
            ]
          }
        }
      }
    }
  ]
}