POST

/v2/writer-detector/{scanId}/check

Use Copyleaks writer detector to differentiate between human texts and AI written texts.

This endpoint will receive submitted text to be checked. At the end of the processing stage, the result will be shown as classifications. Text classification is divided into sections. Each section may have a different classification.

Need plagiarism scan?

If you are scanning your contents for Plagiarism too, we recommend to use the Plagiarism API instead. While using the Plagiarism API, you can enable writer detector by using the properties.aiGeneratedText.detect flag.

You need to login with a user and api key in order to access this method.

Add this HTTP header to your request:

Authorization: Bearer <Your-Login-Token>

Not sure how to generate your login token? Read here .

Request

URL Parameters

Name

Description

scanIdREQUIRED

A unique scan id provided by you.

We recommend you use the same id in your database to represent the scan in the Copyleaks database. This will help you to debug incidents.

Using the same ID for the same file will help you to avoid network problems that may lead to multiple scans for the same file.

String

Length: 3-36 characters.

Allowed characters are [a-z0-9] and the following symbols: !@$^&-+%=_(){}<>';:/.",~`|

learn more about the criteria for creating a Scan ID .

Body Parameters

Name

Description

textREQUIRED

A text string.

Supported Languages:

English
Spanish
French
Portuguese
German
Italian
Russian
Polish
Romanian
Dutch
Swedish
Czech
Norwegian
Korean
Japanese
Chinese

String

Max length: 25000 characters.

Min length: 255 characters.

sandbox

Use sandbox mode to test your integration with the Copyleaks API for free.

Submit content for an AI detection and get returned mock results, simulating Copyleaks's API functionality to ensure you have successfully integrated the API.

This feature is intended to be used for development purposes only.

Boolean

Default: false

Request Example

RAW

cURL

Python

Node.js

PHP

POST

https://api.copyleaks.com/v2/writer-detector/my-custom-id/check

Headers

Body

text: "Copyleaks is a comprehensive plagiarism detection platform that performs extensive searches across 60 trillion websites, 15,000+ academic journals, 20+ code data repositories, and 1M+ internal documents. Using AI-powered text analysis, easily scan documents, raw text, code, and URLs and instantly receive detailed reporting on the findings."

Response

Description

Name

Description

results

An array of all the classifications. The sections contain their position, classification and the probability for that classification.

Object Array

results.classification

Indicates the text classification.

Integer (Enum)

Available Options:
1 - Human
2 - AI Generated Text

results.probability

The classification confidence.

Float

results.matches

Provides information about the match objects that have been found in the text.

Object Array

results.matches.text

Describes the text positions for the result (textual version).

Object

results.matches.text.chars

Positions of the characters in that sepcific result.

Object

results.matches.text.chars.starts

Character match start positions.

Integer Array

results.matches.text.chars.lengths

Character match lengths.

Integer Array

results.matches.text.words

Positions of the words in that sepcific result.

Object

results.matches.text.words.starts

Words match start positions.

Integer Array

results.matches.text.words.lengths

Words match lengths.

Integer Array

summary

Aggregated detection results.

Object

summary.human

Indicates the confidence in which this text was written by a human.

Float

summary.ai

Indicates the confidence in which this text was written by AI.

Float

scannedDocument

General information about the scanned document.

Object

scannedDocument.scanId

The scan id given by the user.

String

scannedDocument.totalWords

Total number of words found in the scanned text.

Integer

scannedDocument.totalExcluded

Total excluded words from the text.

Integer

scannedDocument.actualCredits

The cost of credits for this scan.

Integer

scannedDocument.expectedCredits

The amount of credits that was expected to be spent on the scan.

Integer

scannedDocument.creationTime

Creation date of the scan.

String (datetime)

Time zone: UTC+0

Codes

Status Code

Description

Example

200

The command was executed.

{
  "results": [
    {
      "classification": 2,
      "probability": 1,
      "matches": [
        {
          "text": {
            "chars": {
              "starts": [
                0
              ],
              "lengths": [
                1509
              ]
            },
            "words": {
              "starts": [
                0
              ],
              "lengths": [
                221
              ]
            }
          }
        }
      ]
    }
  ],
  "summary": {
    "human": 0,
    "ai": 1
  },
  "scannedDocument": {
    "scanId": "check",
    "totalWords": 221,
    "totalExcluded": 0,
    "actualCredits": 1,
    "expectedCredits": 1,
    "creationTime": "2023-01-10T10:07:58.9459512Z"
  }
}

See resource description

400

Bad request.

{
  "errorCode": "unsupported-lang",
  "errorMessage": "Unsupported language 'he' for AI text detection"
}

401

Unauthorized

Authorization has been denied for this request.

429

Too many requests have been sent. The request has been rejected.

TIP: Reuse your key as much as possible.

Other resources:

Performance Considerations ^Important! - How to improve your scan performance.

Exponential Backoff - Algorithm that helps applications define a retry strategy for consuming a network service.

Do you have a technical question?

Use stackoverflow.com to get help from our development team and other Copyleaks users.

Ask a Question