openapi: 3.0.0
info:
  description: Context aware, pluggable and customizable PII anonymization service for text and images.
  version: "2.0"
  title: Presidio
  contact:
    name: Presidio support
    email: presidio@microsoft.com
    url: https://github.com/microsoft/presidio
  x-logo:
    url: "https://upload.wikimedia.org/wikipedia/commons/4/44/Microsoft_logo.svg"
  license:
    name: MIT
    url: 'https://github.com/microsoft/presidio/blob/main/LICENSE'
externalDocs:
  description: Presidio documentation.
  url: 'https://microsoft.github.io/presidio/'
tags:
  - name: Analyzer
    description: Detecting PII entities in text
  - name: Anonymizer
    description: Anonymizing detected PII text entities with desired values.

paths:

  /analyze:
    post:
      servers:
        - url: https://presidio-analyzer-prod.azurewebsites.net
      tags:
        - Analyzer
      summary: "Analyze Text"
      description: "Recognizes PII entities in a given text and returns their types, locations and score"
      requestBody:
        $ref: "#/components/requestBodies/AnalyzeRequest"

      responses:
        200:
          description: OK
          content:
            application/json:
              schema:
                description: "A list analysis results"
                type: array
                items:
                  $ref: "#/components/schemas/RecognizerResultWithAnaysisExplanation"

              examples:
                Enhanced response:
                  value:
                    [
                      { "entity_type": "PERSON", "start": 0, "end": 10, "score": 0.85,
                        "analysis_explanation": {
                          "recognizer": "SpacyRecognizer", "pattern_name": null, "pattern": null, "original_score": 0.85,
                          "score": 0.85, "textual_explanation": "Identified as PERSON by Spacy's Named Entity Recognition",
                          "score_context_improvement": 0, "supportive_context_word": "", "validation_result": null
                        },
                        "recognition_metadata": {
                          "recognizer_name": "SpacyRecognizer"
                        }
                      },
                      { "entity_type": "US_DRIVER_LICENSE", "start": 30, "end": 38, "score": 0.6499999999999999,
                        "analysis_explanation": {
                          "recognizer": "UsLicenseRecognizer", "pattern_name": "Driver License - Alphanumeric (weak)",
                          "pattern": "\\\\b([A-Z][0-9]{3,6}|[A-Z][0-9]{5,9}|[A-Z][0-9]{6,8}|[A-Z][0-9]{4,8}|[A-Z][0-9]{9,11}|[A-Z]{1,2}[0-9]{5,6}|H[0-9]{8}|V[0-9]{6}|X[0-9]{8}|A-Z]{2}[0-9]{2,5}|[A-Z]{2}[0-9]{3,7}|[0-9]{2}[A-Z]{3}[0-9]{5,6}|[A-Z][0-9]{13,14}|[A-Z][0-9]{18}|[A-Z][0-9]{6}R|[A-Z][0-9]{9}|[A-Z][0-9]{1,12}|[0-9]{9}[A-Z]|[A-Z]{2}[0-9]{6}[A-Z]|[0-9]{8}[A-Z]{2}|[0-9]{3}[A-Z]{2}[0-9]{4}|[A-Z][0-9][A-Z][0-9][A-Z]|[0-9]{7,8}[A-Z])\\\\b",
                          "original_score": 0.3, "score": 0.6499999999999999, "textual_explanation": null,
                          "score_context_improvement": 0.3499999999999999, "supportive_context_word": "driver",
                          "validation_result": null
                        },
                        "recognition_metadata": {
                          "recognizer_name": "UsLicenseRecognizer"
                        }
                      }
                    ]
                Lean response:
                  value:
                    [
                      {
                        "analysis_explanation": null,
                        "end": 38,
                        "entity_type": "US_DRIVER_LICENSE",
                        "score": 0.6499999999999999,
                        "start": 30,
                        "recognition_metadata": {
                          "recognizer_name": "UsLicenseRecognizer"
                        }
                      }
                    ]

  /recognizers:
    get:
      servers:
        - url: https://presidio-analyzer-prod.azurewebsites.net
      tags:
        - Analyzer
      summary: "Get Recognizers"
      description: "Get the available PII recognizers for a given language"
      parameters:
        - in: query
          name: language
          schema:
            type: string
            example: en
          description: "Two characters for the desired language in ISO_639-1 format"
      responses:
        200:
          description: OK
          content:
            application/json:
              schema:
                description: "A list of supported recognizers"
                type: array
                items:
                  type: string
                  description: "Recognizer name"
              example:
                [
                    "CryptoRecognizer", "CreditCardRecognizer", "IbanRecognizer", "UsPhoneRecognizer",
                    "EmailRecognizer","UsPassportRecognizer", "NhsRecognizer", "IpRecognizer",
                    "SpacyRecognizer","SgFinRecognizer", "UsSsnRecognizer","UsBankRecognizer",
                    "DomainRecognizer", "UsLicenseRecognizer", "UsItinRecognizer"
                ]

  /supportedentities:
    get:
      servers:
        - url: https://presidio-analyzer-prod.azurewebsites.net
      tags:
        - Analyzer
      summary: "Get supported entities"
      description: "Get the list of PII entities Presidio-Analyzer is capable of detecting"
      parameters:
        - in: query
          name: language
          schema:
            type: string
            example: en
          description: "Two characters for the desired language in ISO_639-1 format"
      responses:
        200:
          description: OK
          content:
            application/json:
              schema:
                description: "A list of supported entities"
                type: array
                items:
                  $ref: "#/components/schemas/EntityTypes"
              example:
                [ "PHONE_NUMBER", "US_DRIVER_LICENSE", "US_PASSPORT", "LOCATION", "CREDIT_CARD", "CRYPTO",
                  "UK_NHS", "US_SSN", "US_BANK_NUMBER", "EMAIL_ADDRESS", "DATE_TIME", "IP_ADDRESS", "PERSON", "IBAN_CODE",
                  "NRP", "US_ITIN", "MEDICAL_LICENSE", "URL" ]

  /anonymize:
    post:
      servers:
        - url: https://presidio-anonymizer-prod.azurewebsites.net
      tags:
        - Anonymizer
      summary: "Anonymize Text"
      requestBody:
        $ref: "#/components/requestBodies/AnonymizeRequest"

      responses:
        200:
          description: OK
          content:
            application/json:
              schema:
                $ref: "#/components/schemas/AnonymizeResponse"
              examples:
                Replace and Redact Anonymizers:
                  value:
                    { "text": "hello world, my name is ANONYMIZED. My number is: ", "items": [ { "operator": "redact", "entity_type": "PHONE_NUMBER", "start": 50, "end": 50, "text": "" }, { "operator": "replace", "entity_type": "NAME", "start": 24, "end": 34, "text": "ANONYMIZED" } ] }
                Replace as default Anonymizer:
                  value:
                    { "text": "hello world, my name is ANONYMIZED. My number is: ANONYMIZED", "items": [ { "operator": "replace", "entity_type": "PHONE_NUMBER", "start": 50, "end": 60, "text": "ANONYMIZED" }, { "operator": "replace", "entity_type": "NAME", "start": 24, "end": 34, "text": "ANONYMIZED" } ] }
        400:
          $ref: "#/components/responses/400BadRequest"

        422:
          $ref: "#/components/responses/422UnprocessableEntity"

  /anonymizers:
    get:
      servers:
        - url: https://presidio-anonymizer-prod.azurewebsites.net
      tags:
        - Anonymizer
      summary: "Get supported anonymizers"

      responses:
        200:
          description: OK
          content:
            application/json:
              schema:
                description: "A list of all built-in supported anonymizers"
                type: array
                items:
                  description: "The Anonymizer name"
                  type: string
                  example: "mask"
              example:
                [ "hash", "mask", "redact", "replace", "encrypt" ]


  /deanonymize:
    post:
      servers:
        - url: https://presidio-anonymizer-prod.azurewebsites.net
      tags:
        - Anonymizer
      summary: "Deanonymize Text"
      requestBody:
        $ref: "#/components/requestBodies/DeanonymizeRequest"

      responses:
        200:
          description: OK
          content:
            application/json:
              schema:
                $ref: "#/components/schemas/DeanonymizeResponse"
              examples:
                Decrypt Single PII:
                  value:
                    { "text": "text_for_encryption", "items": [ { "start": 0, "end": 19, "operator":"decrypt", "text": "text_for_encryption","entity_type": "NUMBER" } ] }
        400:
          $ref: "#/components/responses/400BadRequest"

        422:
          $ref: "#/components/responses/422UnprocessableEntity"

  /deanonymizers:
    get:
      servers:
        - url: https://presidio-anonymizer-prod.azurewebsites.net
      tags:
        - Anonymizer
      summary: "Get supported deanonymizers"

      responses:
        200:
          description: OK
          content:
            application/json:
              schema:
                description: "A list of all built-in supported deanonymizers"
                type: array
                items:
                  description: "The Deanonymizer name"
                  type: string
                  example: "decrypt"
              example:
                [ "decrypt" ]

  /health:
    get:
      servers:
        - url: https://presidio-anonymizer-prod.azurewebsites.net
      tags:
        - Anonymizer
        - Analyzer
      summary: "Healthcheck"
      responses:
        200:
          description: OK
          content:
            text/plain:
              schema:
                type: string
                example: Presidio Anonymizer service is up

components:
  requestBodies:
    AnalyzeRequest:
      required: true
      content:
        application/json:
          schema:
            $ref: "#/components/schemas/AnalyzeRequest"
          examples:
            Minimal Request:
              value:
                {
                  "text": "John Smith drivers license is AC432223",
                  "language": "en"
                }
            Enhanced Request :
              value:
                {
                  "text": "John Smith drivers license is AC432223 and the zip code is 12345",
                  "language": "en",
                  "return_decision_process": false,
                  "correlation_id": "123e4567-e89b-12d3-a456-426614174000",
                  "score_threshold": 0.6,
                  "entities": ["US_DRIVER_LICENSE", "ZIP"],
                  "trace": false,
                  "ad_hoc_recognizers":[
                    {
                    "name": "Zip code Recognizer",
                    "supported_language": "en",
                    "patterns": [
                        {
                        "name": "zip code (weak)", 
                        "regex": "(\\b\\d{5}(?:\\-\\d{4})?\\b)", 
                        "score": 0.01
                        }
                    ],
                    "context": ["zip", "code"],
                    "supported_entity":"ZIP"
                    }
                  ]
                }

    AnonymizeRequest:
      required: true
      content:
        application/json:
          schema:
            $ref: "#/components/schemas/AnonymizeRequest"
          examples:
            Replace and Redact Anonymizers:
              value:
                {
                  "text": "hello world, my name is Jane Doe. My number is: 034453334",
                  "anonymizers": {
                    "PERSON": { "type": "redact" },
                    "PHONE_NUMBER": { "type": "replace", "new_value": "ANONYMIZED" }
                  },
                  "analyzer_results": [
                    { "start": 24, "end": 32, "score": 0.8, "entity_type": "PERSON" },
                    { "start": 48, "end": 57,  "score": 0.95, "entity_type": "PHONE_NUMBER" }
                  ]
                }
            Replace as default Anonymizer:
              value:
                {
                  "text": "hello world, my name is Jane Doe.",
                  "anonymizers": {
                    "DEFAULT": { "type": "replace", "new_value": "ANONYMIZED" },
                  },
                  "analyzer_results": [
                    { "start": 24, "end": 32, "score": 0.8, "entity_type": "PERSON" },
                  ]
                }

    DeanonymizeRequest:
      required: true
      content:
        application/json:
          schema:
            $ref: "#/components/schemas/DeanonymizeRequest"
          examples:
            Deanonymize text:
              value:
                {
                  "text": "My name is S184CMt9Drj7QaKQ21JTrpYzghnboTF9pn/neN8JME0=",
                  "deanonymizers": {
                    "PERSON": {
                      "type": "decrypt",
                      "key": "WmZq4t7w!z%C&F)J"
                    }
                  },
                  "anonymizer_results": [ {
                    "start": 11,
                    "end": 55,
                    "entity_type": "PERSON"
                  } ]
                }


  schemas:
    AnalyzeRequest:
      type: object
      required:
        - text
        - language
      properties:
        text:
          type: string
          description: "The text to analyze"
          example: "hello world, my name is Jane Doe. My number is: 034453334"
        language:
          type: string
          description: "Two characters for the desired language in ISO_639-1 format"
          example: "en"
        correlation_id:
          type: string
          description: "A correlation id to append to headers and traces"
        score_threshold:
          type: number
          format: double
          description: "The minimal detection score threshold"
        entities:
          type: array
          items:
            $ref: "#/components/schemas/EntityTypes"
          description: "A list of entities to analyze"
        return_decision_process:
          type: boolean
          description: "Whether to include analysis explanation in the response"
        ad_hoc_recognizers:
          type: array
          description: "list of recognizers to be used in the context of this request only (ad-hoc)."
          items:
            $ref: "#/components/schemas/PatternRecognizer"
        context:
          type: array
          description: "list of context words which may help to raise recognized entities confidence"
          items:
            description: "The context word"
            type: string
            example: "address"

    AnonymizeRequest:
      type: object
      required:
        - text
        - analyzer_results
      properties:
        text:
          type: string
          description: "The text to anonymize"
          example: "hello world, my name is Jane Doe. My number is: 034453334"
        anonymizers:
          description: "Object where the key is DEFAULT or the ENTITY_TYPE and the value is the anonymizer definition"
          type: object
          additionalProperties:
            anyOf:
              - $ref: "#/components/schemas/Replace"
              - $ref: "#/components/schemas/Redact"
              - $ref: "#/components/schemas/Mask"
              - $ref: "#/components/schemas/Hash"
              - $ref: "#/components/schemas/Encrypt"
          default:
            { "DEFAULT": { "type": "replace", "new_value": "<ENTITY_TYPE>" } }

        analyzer_results:
          type: array
          description: "Array of analyzer detections"
          items:
            $ref: "#/components/schemas/RecognizerResult"

    DeanonymizeRequest:
      type: object
      required:
        - text
        - anonymizer_results
        - deanonymizers
      properties:
        text:
          type: string
          description: "The anonymized text"
          example: "My name is S184CMt9Drj7QaKQ21JTrpYzghnboTF9pn/neN8JME0="
        deanonymizers:
          description: "Object where the key is DEFAULT or the ENTITY_TYPE and the value is decrypt since it is the only one supported"
          type: object
          additionalProperties:
            anyOf:
              - $ref: "#/components/schemas/Decrypt"
          default:
            { "DEFAULT": { "type": "decrypt", "key": "3t6w9z$C&F)J@NcR" } }
        anonymizer_results:
          type: array
          description: "Array of anonymized PIIs"
          items:
            $ref: "#/components/schemas/OperatorResult"


    RecognizerResult:
      type: object
      required:
        - start
        - end
        - score
        - entity_type
      properties:
        start:
          type: integer
          description: "Where the PII starts"
          example: 24
        end:
          type: integer
          description: "Where the PII ends"
          example: 32
        score:
          type: number
          format: double
          description: "The PII detection score"
          example: 0.8
        entity_type:
          $ref: "#/components/schemas/EntityTypes"
        recognition_metadata:
          type: object
          $ref: "#/components/schemas/RecognizedMetadata"

    RecognizedMetadata:
      type: object
      properties:
        recognizer_name:
          type: string
          description: "Name of recognizer that made the decision"

    RecognizerResultWithAnaysisExplanation:
      allOf:
        - $ref: '#/components/schemas/RecognizerResult'
        - type: object
          properties:
            analysis_explanation:
              $ref: "#/components/schemas/AnalysisExplanation"

    AnalysisExplanation:
      type: object
      properties:
        recognizer:
          type: string
          description: "Name of recognizer that made the decision"
        pattern_name:
          type: string
          description: "name of pattern (if decision was made by a PatternRecognizer)"
        pattern:
          type: string
          description: "Regex pattern that was applied (if PatternRecognizer)"
        original_score:
          type: number
          format: double
          description: "Recognizer's confidence in result"
        score:
          type: number
          format: double
          description: "The PII detection score"
        textual_explanation:
          type: string
          description: "Free text for describing a decision of a logic or model"
        score_context_improvement:
          type: number
          format: double
          description: "Difference from the original score"
        supportive_context_word:
          type: string
          description: "The context word which helped increase the score"
        validation_result:
          type: number
          format: double
          description: "Result of a validation (e.g. checksum)"
    
    Pattern:
      type: object
      properties:
        name:
          type: string
          description: "Name of regular expression pattern"
        regex:
          type: string
          description: "Regex pattern string"
        score:
          type: number
          format: double
          description: "Detection confidence of this pattern (0.01 if very noisy, 0.6-1.0 if very specific)"


    PatternRecognizer:
      type: object
      description: "A regular expressions or deny-list based recognizer"
      properties:
        name:
          type: string
          description: "Name of recognizer"
        supported_language:
          type: string
          description: "Language code supported by this recognizer"
        patterns:
          description: "List of type Pattern containing regex expressions with additional metadata."
          type: array
          items:
            $ref: "#/components/schemas/Pattern"
        deny_list:
          type: array
          description: "List of words to be returned as PII if found."
          items:
            type: string
        context:
          description: "List of words to be used to increase confidence if found in the vicinity of detected entities."
          type: array
          items:
            type: string
        supported_entity:
          type: string
          description: "The name of entity this ad hoc recognizer detects"

    EntityTypes:
      description: "The supported PII entity types."
      type: string
      example: PERSON

    Replace:
      title: Replace
      description: "Replace with a given value"
      type: object
      required:
        - type
        - new_value
      properties:
        type:
          type: string
          description: "replace"
          example: replace
        new_value:
          type: string
          description: "The string to replace with"
          example: VALUE

    Redact:
      title: Redact
      description: "Replace with an empty string"
      type: object
      required:
        - type
      properties:
        type:
          type: string
          description: "redact"
          example: redact

    Mask:
      title: Mask
      description: "Replace with a given character"
      type: object
      required:
        - type
        - masking_char
        - chars_to_mask
      properties:
        type:
          type: string
          description: "mask"
          example: mask
        masking_char:
          type: string
          description: "The replacement character"
          example: "*"
        chars_to_mask:
          type: integer
          description: "The amount of characters that should be replaced"
          example: 4
        from_end:
          type: boolean
          description: "Whether to mask the PII from it's end"
          example: true
          default: false

    Hash:
      title: Hash
      description: "Replace with hashed value"
      type: object
      required:
        - type
      properties:
        type:
          type: string
          description: "hash"
          example: hash
        hash_type:
          type: string
          description: "The hashing algorithm"
          enum:
            - md5
            - sha256
            - sha512
          example: md5
          default: md5

    Encrypt:
      title: Encrypt
      description: "Replace with an encrypted value"
      type: object
      required:
        - type
        - key
      properties:
        type:
          type: string
          description: "encrypt"
          example: encrypt
        key:
          type: string
          description: "Cryptographic key of length 128, 192 or 256 bits, in a string format"
          example: "3t6w9z$C&F)J@NcR"

    Decrypt:
      title: Decrypt
      description: "Replace encrypted PII decrypted text"
      type: object
      required:
        - type
        - key
      properties:
        type:
          type: string
          description: "decrypt"
          example: decrypt
        key:
          type: string
          description: "Cryptographic key of length 128, 192 or 256 bits, in a string format"
          example: "3t6w9z$C&F)J@NcR"

    AnonymizeResponse:
      type: object
      properties:
        text:
          type: string
        items:
          type: array
          description: "Array of anonymized entities"
          items:
            $ref: "#/components/schemas/OperatorResult"

    OperatorResult:
      required:
        - start
        - end
        - entity_type
      type: object
      properties:
        operator:
          type: string
          description: "Name of the used operator"
        entity_type:
          type: string
          description: "Type of the PII entity"
        start:
          type: integer
          description: "Start index of the changed text"
        end:
          type: integer
          description: "End index in the changed text"
        text:
          type: string
          description: "The new text returned"

    DeanonymizeResponse:
      type: object
      properties:
        text:
          type: string
        items:
          type: array
          description: "Array of deanonymized entities"
          items:
            $ref: "#/components/schemas/OperatorResult"

  responses:
    400BadRequest:
      description: Bad request
      content:
        application/json:
          schema:
            type: object
            properties:
              error:
                type: string
                example: "Invalid request json"

    422UnprocessableEntity:
      description: Unprocessable Entity
      content:
        application/json:
          schema:
            type: object
            properties:
              error:
                type: string
                example: "Invalid input, text can not be empty"