fil/docs/schemas/document-content.schema.json

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://kreuzberg.dev/schemas/document-content.schema.json",
  "title": "JsonDocument",
  "description": "Kreuzberg structured JSON content output. A heading-driven tree where headings create nested sections.",
  "type": "object",
  "properties": {
    "title": {
      "type": "string",
      "description": "Document title, if found."
    },
    "body": {
      "type": "array",
      "items": {
        "$ref": "#/$defs/node"
      },
      "description": "Body content nodes."
    }
  },
  "required": ["body"],
  "additionalProperties": false,
  "$defs": {
    "node": {
      "description": "A node in the JSON document tree, discriminated by the 'type' field.",
      "oneOf": [
        { "$ref": "#/$defs/section" },
        { "$ref": "#/$defs/paragraph" },
        { "$ref": "#/$defs/table" },
        { "$ref": "#/$defs/code" },
        { "$ref": "#/$defs/formula" },
        { "$ref": "#/$defs/list" },
        { "$ref": "#/$defs/image" },
        { "$ref": "#/$defs/blockquote" }
      ]
    },
    "section": {
      "type": "object",
      "description": "A heading-delimited section containing nested content.",
      "properties": {
        "type": {
          "const": "section"
        },
        "heading": {
          "type": "string",
          "description": "The heading text for this section."
        },
        "level": {
          "type": "integer",
          "minimum": 1,
          "maximum": 6,
          "description": "Heading level (1-6)."
        },
        "body": {
          "type": "array",
          "items": {
            "$ref": "#/$defs/node"
          },
          "description": "Child content nodes within this section."
        }
      },
      "required": ["type", "heading", "level", "body"],
      "additionalProperties": false
    },
    "paragraph": {
      "type": "object",
      "description": "A text paragraph.",
      "properties": {
        "type": {
          "const": "paragraph"
        },
        "text": {
          "type": "string",
          "description": "The paragraph text content."
        }
      },
      "required": ["type", "text"],
      "additionalProperties": false
    },
    "table": {
      "type": "object",
      "description": "A table with headers and rows.",
      "properties": {
        "type": {
          "const": "table"
        },
        "headers": {
          "type": "array",
          "items": {
            "type": "string"
          },
          "description": "Column header labels."
        },
        "rows": {
          "type": "array",
          "items": {
            "type": "array",
            "items": {
              "type": "string"
            }
          },
          "description": "Table data rows (each row is an array of cell strings)."
        },
        "caption": {
          "type": "string",
          "description": "Optional table caption."
        }
      },
      "required": ["type", "headers", "rows"],
      "additionalProperties": false
    },
    "code": {
      "type": "object",
      "description": "A code block with optional language.",
      "properties": {
        "type": {
          "const": "code"
        },
        "text": {
          "type": "string",
          "description": "The code content."
        },
        "language": {
          "type": "string",
          "description": "Programming language identifier."
        }
      },
      "required": ["type", "text"],
      "additionalProperties": false
    },
    "formula": {
      "type": "object",
      "description": "A mathematical formula.",
      "properties": {
        "type": {
          "const": "formula"
        },
        "text": {
          "type": "string",
          "description": "The formula text (e.g. LaTeX notation)."
        }
      },
      "required": ["type", "text"],
      "additionalProperties": false
    },
    "list": {
      "type": "object",
      "description": "An ordered or unordered list.",
      "properties": {
        "type": {
          "const": "list"
        },
        "ordered": {
          "type": "boolean",
          "description": "Whether the list is ordered (numbered) or unordered (bulleted)."
        },
        "items": {
          "type": "array",
          "items": {
            "type": "string"
          },
          "description": "List item text content."
        }
      },
      "required": ["type", "ordered", "items"],
      "additionalProperties": false
    },
    "image": {
      "type": "object",
      "description": "An image reference.",
      "properties": {
        "type": {
          "const": "image"
        },
        "alt": {
          "type": "string",
          "description": "Alternative text description of the image."
        },
        "src": {
          "type": "string",
          "description": "Image source path or generated filename."
        }
      },
      "required": ["type"],
      "additionalProperties": false
    },
    "blockquote": {
      "type": "object",
      "description": "A blockquote containing nested content.",
      "properties": {
        "type": {
          "const": "blockquote"
        },
        "body": {
          "type": "array",
          "items": {
            "$ref": "#/$defs/node"
          },
          "description": "Child content nodes within this blockquote."
        }
      },
      "required": ["type", "body"],
      "additionalProperties": false
    }
  }
}