Reducto Classify API

openapi: 3.1.0
info:
  title: Reducto Classify API
  version: 1.0.0
  description: Classify documents into a defined set of categories and run citation lookups against parsed content.
  contact:
    name: Reducto Support
    email: [email protected]
    url: https://reducto.ai/contact
  license:
    name: Reducto Terms of Service
    url: https://reducto.ai/terms
servers:
- url: https://platform.reducto.ai
  description: Reducto production platform
security:
- SkippableHTTPBearer: []
tags:
- name: Classify
paths:
  /classify:
    post:
      summary: Classify
      operationId: classify_classify_post
      security:
      - SkippableHTTPBearer: []
      parameters:
      - name: user-id
        in: header
        required: false
        schema:
          anyOf:
          - type: string
          - type: 'null'
          title: User-Id
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ClassifyConfig'
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ClassifyResponse'
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
      tags:
      - Classify
  /cite:
    post:
      summary: Cite
      description: 'Locate exact bounding boxes for text citations in a parsed document.


        Requires the document to have been parsed with return_ocr_data=true.'
      operationId: cite_cite_post
      security:
      - SkippableHTTPBearer: []
      parameters:
      - name: user-id
        in: header
        required: false
        schema:
          anyOf:
          - type: string
          - type: 'null'
          title: User-Id
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CiteRequest'
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/CiteResponse'
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
      tags:
      - Classify
components:
  schemas:
    ClassifyResponse:
      properties:
        response_type:
          type: string
          const: classify
          title: Response Type
          default: classify
        job_id:
          type: string
          title: Job Id
        result:
          $ref: '#/components/schemas/ClassifyResponseCategory'
        response_confidence:
          anyOf:
          - $ref: '#/components/schemas/ResponseConfidence'
          - type: 'null'
        duration:
          anyOf:
          - type: number
          - type: 'null'
          title: Duration
          description: The duration of the classify request in seconds.
      type: object
      required:
      - job_id
      - result
      title: ClassifyResponse
      description: Response from classify job - returned when polling /job/{job_id}
    CitationQuery:
      properties:
        text:
          type: string
          minLength: 1
          title: Text
          description: Text to locate. Whitespace is normalized for matching.
        bbox_filter:
          anyOf:
          - $ref: '#/components/schemas/BoundingBox'
          - type: 'null'
          description: Optional region to limit search.
      type: object
      required:
      - text
      title: CitationQuery
      description: A text to locate in the document.
    ValidationError:
      properties:
        loc:
          items:
            anyOf:
            - type: string
            - type: integer
          type: array
          title: Location
        msg:
          type: string
          title: Message
        type:
          type: string
          title: Error Type
        input:
          title: Input
        ctx:
          type: object
          title: Context
      type: object
      required:
      - loc
      - msg
      - type
      title: ValidationError
    ParseChunk-Input:
      properties:
        content:
          type: string
          title: Content
          description: The content of the chunk extracted from the document.
        embed:
          type: string
          title: Embed
          description: Chunk content optimized for embedding and retrieval.
        enriched:
          anyOf:
          - type: string
          - type: 'null'
          title: Enriched
          description: The enriched content of the chunk extracted from the document.
        enrichment_success:
          type: boolean
          title: Enrichment Success
          description: Whether the enrichment was successful.
          default: false
        blocks:
          items:
            $ref: '#/components/schemas/ParseBlock-Input'
          type: array
          title: Blocks
      type: object
      required:
      - content
      - embed
      - enriched
      - blocks
      title: ParseChunk
    CiteResponse:
      properties:
        results:
          items:
            $ref: '#/components/schemas/CitationQueryResult'
          type: array
          title: Results
          description: Results in same order as input queries (1:1 correspondence).
        duration:
          type: number
          title: Duration
          description: Processing time in seconds.
      type: object
      required:
      - results
      - duration
      title: CiteResponse
      description: Citation location results.
    ResponseConfidence:
      properties:
        categories:
          items:
            $ref: '#/components/schemas/CategoryConfidence'
          type: array
          title: Categories
      type: object
      required:
      - categories
      title: ResponseConfidence
      description: Overall confidence breakdown for classification response.
    FullResult-Input:
      properties:
        type:
          type: string
          const: full
          title: Type
          description: type = 'full'
        chunks:
          items:
            $ref: '#/components/schemas/ParseChunk-Input'
          type: array
          title: Chunks
        ocr:
          anyOf:
          - $ref: '#/components/schemas/OCRResult-Input'
          - type: 'null'
        custom:
          anyOf:
          - {}
          - type: 'null'
          title: Custom
      type: object
      required:
      - type
      - chunks
      title: FullResult
    ClassifyResponseCategory:
      properties:
        category:
          type: string
          title: Category
      type: object
      required:
      - category
      title: ClassifyResponseCategory
    CriteriaConfidence:
      properties:
        criterion:
          type: string
          title: Criterion
        confidence:
          type: string
          enum:
          - high
          - low
          title: Confidence
      type: object
      required:
      - criterion
      - confidence
      title: CriteriaConfidence
      description: Confidence result for a single criterion.
    BoundingBox:
      properties:
        left:
          type: number
          title: Left
        top:
          type: number
          title: Top
        width:
          type: number
          title: Width
        height:
          type: number
          title: Height
        page:
          type: integer
          title: Page
          description: The page number of the bounding box (1-indexed).
        original_page:
          type: integer
          title: Original Page
          description: The page number in the original document of the bounding box (1-indexed).
      type: object
      required:
      - left
      - top
      - width
      - height
      - page
      title: BoundingBox
    CategoryConfidence:
      properties:
        category:
          type: string
          title: Category
        confidence:
          type: number
          title: Confidence
        criteria_confidence:
          items:
            $ref: '#/components/schemas/CriteriaConfidence'
          type: array
          title: Criteria Confidence
      type: object
      required:
      - category
      - confidence
      - criteria_confidence
      title: CategoryConfidence
      description: Confidence result for a category.
    CitationMatch:
      properties:
        page:
          type: integer
          title: Page
          description: Page number (1-indexed).
        bboxes:
          items:
            $ref: '#/components/schemas/BoundingBox'
          type: array
          title: Bboxes
          description: Bounding boxes for the match. Multiple boxes for multi-line text.
      type: object
      required:
      - page
      - bboxes
      title: CitationMatch
      description: A location where the citation text was found.
    GranularConfidence:
      properties:
        extract_confidence:
          anyOf:
          - type: number
          - type: 'null'
          title: Extract Confidence
        parse_confidence:
          anyOf:
          - type: number
          - type: 'null'
          title: Parse Confidence
      type: object
      title: GranularConfidence
    PageRange:
      properties:
        start:
          anyOf:
          - type: integer
          - type: 'null'
          title: Start
          description: The page number to start processing from (1-indexed).
        end:
          anyOf:
          - type: integer
          - type: 'null'
          title: End
          description: The page number to stop processing at (1-indexed).
      type: object
      title: PageRange
    OCRWord:
      properties:
        text:
          type: string
          title: Text
        bbox:
          $ref: '#/components/schemas/BoundingBox'
        confidence:
          anyOf:
          - type: number
          - type: 'null'
          title: Confidence
          description: OCR confidence score between 0 and 1, where 1 indicates highest confidence
        chunk_index:
          anyOf:
          - type: integer
          - type: 'null'
          title: Chunk Index
          description: The index of the chunk that the word belongs to.
        rotation:
          anyOf:
          - type: integer
          - type: 'null'
          title: Rotation
          description: The rotation angle in degrees, from 0 to 360, counterclockwise.
      type: object
      required:
      - text
      - bbox
      title: OCRWord
    OCRResult-Input:
      properties:
        words:
          items:
            $ref: '#/components/schemas/OCRWord'
          type: array
          title: Words
        lines:
          items:
            $ref: '#/components/schemas/OCRLine'
          type: array
          title: Lines
      type: object
      required:
      - words
      - lines
      title: OCRResult
    UploadResponse:
      properties:
        file_id:
          type: string
          title: File Id
        presigned_url:
          anyOf:
          - type: string
          - type: 'null'
          title: Presigned Url
      type: object
      required:
      - file_id
      title: UploadResponse
    CiteRequest:
      properties:
        source:
          anyOf:
          - type: string
          - $ref: '#/components/schemas/FullResult-Input'
          title: Source
          description: Either 'jobid://<job_id>' string or full parse result object. The parse must have been run with return_ocr_data=true.
        queries:
          items:
            $ref: '#/components/schemas/CitationQuery'
          type: array
          minItems: 1
          title: Queries
          description: List of text citations to locate.
      type: object
      required:
      - source
      - queries
      title: CiteRequest
      description: Request to locate text citations in a parsed document.
    ClassificationCategory:
      properties:
        category:
          type: string
          title: Category
          description: The category name/label that documents will be classified into (e.g., 'invoice', 'contract', 'receipt').
        criteria:
          items:
            type: string
          type: array
          title: Criteria
          description: A list of criteria, keywords, or descriptions that define what characteristics a document must have
            to be classified into this category (e.g., ['contains billing information', 'has itemized charges']).
      type: object
      required:
      - category
      - criteria
      title: ClassificationCategory
      description: A single classification category with its matching criteria.
    ParseBlock-Input:
      properties:
        type:
          type: string
          enum:
          - Header
          - Footer
          - Title
          - Section Header
          - Page Number
          - List Item
          - Figure
          - Table
          - Key Value
          - Text
          - Comment
          - Signature
          title: Type
          description: The type of block extracted from the document.
        bbox:
          $ref: '#/components/schemas/BoundingBox'
          description: The bounding box of the block extracted from the document.
        content:
          type: string
          title: Content
          description: The content of the block extracted from the document.
        image_url:
          anyOf:
          - type: string
          - type: 'null'
          title: Image Url
          description: (Experimental) The URL of the image associated with the block.
        chart_data:
          anyOf:
          - items:
              type: string
            type: array
          - type: 'null'
          title: Chart Data
          description: (Experimental) The URL/link to chart data JSON for figure blocks processed by chart agent.
        confidence:
          anyOf:
          - type: string
          - type: 'null'
          title: Confidence
          description: The confidence for the block. It is either low or high and takes into account factors like OCR and
            table structure
          default: low
        granular_confidence:
          anyOf:
          - $ref: '#/components/schemas/GranularConfidence'
          - type: 'null'
          description: Granular confidence scores for the block. It is a dictionary of confidence scores for the block. The
            confidence scores will not be None if the user has enabled numeric confidence scores.
        extra:
          anyOf:
          - additionalProperties: true
            type: object
          - type: 'null'
          title: Extra
          description: Extra metadata fields for the block. Fields like 'is_chart' will only appear when set to True.
      type: object
      required:
      - type
      - bbox
      - content
      title: ParseBlock
    ClassifyConfig:
      properties:
        persist_results:
          type: boolean
          title: Persist Results
          description: If True, persist the results indefinitely. Defaults to False.
          default: false
        input:
          anyOf:
          - type: string
          - items:
              type: string
            type: array
          - $ref: '#/components/schemas/UploadResponse'
          title: Input
          description: "For parse/split/extract pipelines, the URL of the document to be processed. You can provide one of\
            \ the following:\n            1. A publicly available URL\n            2. A presigned S3 URL\n            3. A\
            \ reducto:// prefixed URL obtained from the /upload endpoint after directly uploading a document\n           \
            \ 4. A jobid:// prefixed URL obtained from a previous /parse invocation\n            5. A list of URLs (for multi-document\
            \ pipelines, V3 API only)\n\n            For edit pipelines, this should be a string containing the edit instructions "
        classification_schema:
          items:
            $ref: '#/components/schemas/ClassificationCategory'
          type: array
          title: Classification Schema
          description: A list of classification categories and their matching criteria.
          default: []
        page_range:
          anyOf:
          - $ref: '#/components/schemas/PageRange'
          - items:
              $ref: '#/components/schemas/PageRange'
            type: array
          - items:
              type: integer
            type: array
          - type: 'null'
          title: Page Range
          description: The page range to process (1-indexed). By default, the first 5 pages are used. If more than 25 pages
            are selected, only the first 25 (after sorting) are used. Only applies to PDFs; ignored for other document types.
        document_metadata:
          anyOf:
          - type: string
          - type: 'null'
          title: Document Metadata
          description: Optional document-level metadata to include in classification prompts.
      type: object
      required:
      - input
      title: ClassifyConfig
    CitationQueryResult:
      properties:
        matches:
          items:
            $ref: '#/components/schemas/CitationMatch'
          type: array
          title: Matches
          description: All locations where the text was found. Empty if no matches.
      type: object
      required:
      - matches
      title: CitationQueryResult
      description: Results for one query.
    OCRLine:
      properties:
        text:
          type: string
          title: Text
        bbox:
          $ref: '#/components/schemas/BoundingBox'
        confidence:
          anyOf:
          - type: number
          - type: 'null'
          title: Confidence
          description: OCR confidence score between 0 and 1, where 1 indicates highest confidence
        chunk_index:
          anyOf:
          - type: integer
          - type: 'null'
          title: Chunk Index
          description: The index of the chunk that the line belongs to.
        rotation:
          anyOf:
          - type: integer
          - type: 'null'
          title: Rotation
          description: The rotation angle in degrees, from 0 to 360, counterclockwise.
      type: object
      required:
      - text
      - bbox
      title: OCRLine
    HTTPValidationError:
      properties:
        detail:
          items:
            $ref: '#/components/schemas/ValidationError'
          type: array
          title: Detail
      type: object
      title: HTTPValidationError
  securitySchemes:
    SkippableHTTPBearer:
      type: http
      scheme: bearer