API

Every kaggle.api.* operation is listed below with parameter shapes and return types from the plugin Zod schemas.

New to Corsair? See API access, authentication, and error handling.

Competitions

downloadFile

competitions.downloadFile Download a single competition data file (rules acceptance required) Risk: read

await corsair.kaggle.api.competitions.downloadFile({});

Input

Name	Type	Required	Description
`id`	`string`	Yes	—
`fileName`	`string`	Yes	—

Output

Name	Type	Required	Description
`contentType`	`string`	Yes	—
`size`	`number`	Yes	—
`dataBase64`	`string`	Yes	—
`fileName`	`string`	No	—

downloadFiles

competitions.downloadFiles Download all competition data files as a zip (rules acceptance required) Risk: read

await corsair.kaggle.api.competitions.downloadFiles({});

Input

Name	Type	Required	Description
`id`	`string`	Yes	—

Output

Name	Type	Required	Description
`contentType`	`string`	Yes	—
`size`	`number`	Yes	—
`dataBase64`	`string`	Yes	—
`fileName`	`string`	No	—

downloadLeaderboard

competitions.downloadLeaderboard Download the competition leaderboard as a zip/CSV (base64) Risk: read

await corsair.kaggle.api.competitions.downloadLeaderboard({});

Input

Name	Type	Required	Description
`id`	`string`	Yes	—

Output

Name	Type	Required	Description
`contentType`	`string`	Yes	—
`size`	`number`	Yes	—
`dataBase64`	`string`	Yes	—
`fileName`	`string`	No	—

generateSubmissionUrl

competitions.generateSubmissionUrl Generate a pre-signed URL and token for uploading a competition submission file Risk: write

await corsair.kaggle.api.competitions.generateSubmissionUrl({});

Input

Name	Type	Required	Description
`competitionName`	`string`	Yes	—
`contentLength`	`number`	Yes	—
`lastModifiedEpochSeconds`	`number`	Yes	—
`fileName`	`string`	Yes	—

Output: object

Output full type

{
}

list

competitions.list List Kaggle competitions with category, group, sort, and search filters Risk: read

await corsair.kaggle.api.competitions.list({});

Input

Name	Type	Required	Description
`group`	`string`	No	—
`category`	`string`	No	—
`sortBy`	`string`	No	—
`page`	`number`	No	—
`search`	`string`	No	—

Output: object

Output full type

{
}[] | {
  results?: {
  }[],
  data?: {
  }[],
  list?: {
  }[],
  totalResults?: number,
  nextPageToken?: string
} | {
}

listFiles

competitions.listFiles List data files available for a Kaggle competition Risk: read

await corsair.kaggle.api.competitions.listFiles({});

Input

Name	Type	Required	Description
`id`	`string`	Yes	—
`pageToken`	`string`	No	—
`pageSize`	`number`	No	—

Output: object

Output full type

{
}[] | {
  results?: {
  }[],
  data?: {
  }[],
  list?: {
  }[],
  totalResults?: number,
  nextPageToken?: string
} | {
}

submit

competitions.submit Finalize a competition submission using a previously uploaded blob file token Risk: write

await corsair.kaggle.api.competitions.submit({});

Input

Name	Type	Required	Description
`id`	`string`	Yes	—
`blobFileTokens`	`string`	Yes	—
`submissionDescription`	`string`	No	—

Output: object

Output full type

{
}

viewLeaderboard

competitions.viewLeaderboard View competition leaderboard rankings and scores Risk: read

await corsair.kaggle.api.competitions.viewLeaderboard({});

Input

Name	Type	Required	Description
`id`	`string`	Yes	—

Output: object

Output full type

{
}

Datasets

create

datasets.create Create a new Kaggle dataset (requires prior file upload tokens and owner username in id) Risk: write

await corsair.kaggle.api.datasets.create({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`slug`	`string`	Yes	—
`title`	`string`	Yes	—
`subtitle`	`string`	No	—
`description`	`string`	No	—
`isPrivate`	`boolean`	No	—
`licenseName`	`string`	No	—
`files`	`object[]`	No	—

files full type

{
}[]

Output: object

Output full type

{
}

createVersion

datasets.createVersion Create a new version of an existing Kaggle dataset after uploading updated files Risk: write

await corsair.kaggle.api.datasets.createVersion({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`datasetSlug`	`string`	Yes	—
`versionNotes`	`string`	No	—
`files`	`object[]`	No	—

files full type

{
}[]

Output: object

Output full type

{
}

download

datasets.download Download all files from a Kaggle dataset as a zip (returned as base64) Risk: read

await corsair.kaggle.api.datasets.download({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`datasetSlug`	`string`	Yes	—
`datasetVersionNumber`	`number`	No	—

Output

Name	Type	Required	Description
`contentType`	`string`	Yes	—
`size`	`number`	Yes	—
`dataBase64`	`string`	Yes	—
`fileName`	`string`	No	—

downloadFile

datasets.downloadFile Download a single file from a Kaggle dataset (returned as base64) Risk: read

await corsair.kaggle.api.datasets.downloadFile({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`datasetSlug`	`string`	Yes	—
`fileName`	`string`	Yes	—
`datasetVersionNumber`	`number`	No	—

Output

Name	Type	Required	Description
`contentType`	`string`	Yes	—
`size`	`number`	Yes	—
`dataBase64`	`string`	Yes	—
`fileName`	`string`	No	—

getMetadata

datasets.getMetadata Get metadata for a Kaggle dataset including title, licenses, and tags Risk: read

await corsair.kaggle.api.datasets.getMetadata({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`datasetSlug`	`string`	Yes	—

Output: object

Output full type

{
}

getStatus

datasets.getStatus Check processing status after dataset create/version (404 expected when already published) Risk: read

await corsair.kaggle.api.datasets.getStatus({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`datasetSlug`	`string`	Yes	—

Output: object

Output full type

{
}

list

datasets.list List Kaggle datasets with optional search, owner, size, and pagination filters Risk: read

await corsair.kaggle.api.datasets.list({});

Input

Name	Type	Required	Description
`search`	`string`	No	—
`user`	`string`	No	—
`sortBy`	`string`	No	—
`size`	`string`	No	—
`fileType`	`string`	No	—
`licenseName`	`string`	No	—
`tagIds`	`string`	No	—
`page`	`number`	No	—
`maxSize`	`number`	No	—
`minSize`	`number`	No	—

Output: object

Output full type

{
}[] | {
  results?: {
  }[],
  data?: {
  }[],
  list?: {
  }[],
  totalResults?: number,
  nextPageToken?: string
} | {
}

listFiles

datasets.listFiles List files in a Kaggle dataset with optional version and pagination Risk: read

await corsair.kaggle.api.datasets.listFiles({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`datasetSlug`	`string`	Yes	—
`datasetVersionNumber`	`number`	No	—
`pageSize`	`number`	No	—
`pageToken`	`string`	No	—

Output: object

Output full type

{
}[] | {
  results?: {
  }[],
  data?: {
  }[],
  list?: {
  }[],
  totalResults?: number,
  nextPageToken?: string
} | {
}

Kernels

downloadOutput

kernels.downloadOutput Download kernel session output files as base64 (fetches signed GCS URLs from GET /kernels/output) Risk: read

await corsair.kaggle.api.kernels.downloadOutput({});

Input

Name	Type	Required	Description
`userName`	`string`	Yes	—
`kernelSlug`	`string`	Yes	—

Output

Name	Type	Required	Description
`files`	`object[]`	Yes	—
`log`	`string`	No	—
`nextPageToken`	`string`	No	—

files full type

{
  contentType: string,
  size: number,
  dataBase64: string,
  fileName?: string
}[]

getStatus

kernels.getStatus Get execution status of a Kaggle kernel Risk: read

await corsair.kaggle.api.kernels.getStatus({});

Input

Name	Type	Required	Description
`userName`	`string`	Yes	—
`kernelSlug`	`string`	Yes	—

Output: object

Output full type

{
}

list

kernels.list List Kaggle kernels (notebooks/scripts) with filters and pagination Risk: read

await corsair.kaggle.api.kernels.list({});

Input

Name	Type	Required	Description
`page`	`number`	No	—
`pageSize`	`number`	No	—
`search`	`string`	No	—
`group`	`string`	No	—
`user`	`string`	No	—
`language`	`string`	No	—
`kernelType`	`string`	No	—
`outputType`	`string`	No	—
`sortBy`	`string`	No	—
`dataset`	`string`	No	—
`competition`	`string`	No	—
`parentKernel`	`string`	No	—

Output: object

Output full type

{
}[] | {
  results?: {
  }[],
  data?: {
  }[],
  list?: {
  }[],
  totalResults?: number,
  nextPageToken?: string
} | {
}

listOutputFiles

kernels.listOutputFiles List output files produced by a kernel run Risk: read

await corsair.kaggle.api.kernels.listOutputFiles({});

Input

Name	Type	Required	Description
`userName`	`string`	Yes	—
`kernelSlug`	`string`	Yes	—
`pageSize`	`number`	No	—
`pageToken`	`string`	No	—

Output: object

Output full type

{
}[] | {
  results?: {
  }[],
  data?: {
  }[],
  list?: {
  }[],
  totalResults?: number,
  nextPageToken?: string
} | {
}

pull

kernels.pull Pull kernel source code and optional metadata from Kaggle Risk: read

await corsair.kaggle.api.kernels.pull({});

Input

Name	Type	Required	Description
`userName`	`string`	Yes	—
`kernelSlug`	`string`	Yes	—
`metadata`	`boolean`	No	—

Output: object

Output full type

{
}

Models

get

models.get Get details for a Kaggle model Risk: read

await corsair.kaggle.api.models.get({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`modelSlug`	`string`	Yes	—

Output: object

Output full type

{
}

getInstance

models.getInstance Get details for a Kaggle model instance (framework variation) Risk: read

await corsair.kaggle.api.models.getInstance({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`modelSlug`	`string`	Yes	—
`framework`	`string`	Yes	—
`instanceSlug`	`string`	Yes	—

Output: object

Output full type

{
}

list

models.list List Kaggle models with optional owner, search, and pagination Risk: read

await corsair.kaggle.api.models.list({});

Input

Name	Type	Required	Description
`search`	`string`	No	—
`owner`	`string`	No	—
`sortBy`	`string`	No	—
`pageSize`	`number`	No	—
`pageToken`	`string`	No	—

Output: object

Output full type

{
}[] | {
  results?: {
  }[],
  data?: {
  }[],
  list?: {
  }[],
  totalResults?: number,
  nextPageToken?: string
} | {
}

listInstanceVersionFiles

models.listInstanceVersionFiles List files for a specific version of a model instance/variation Risk: read

await corsair.kaggle.api.models.listInstanceVersionFiles({});

Input

Name	Type	Required	Description
`ownerSlug`	`string`	Yes	—
`modelSlug`	`string`	Yes	—
`framework`	`string`	Yes	—
`instanceSlug`	`string`	Yes	—
`versionNumber`	`number`	Yes	—
`pageSize`	`number`	No	—
`pageToken`	`string`	No	—

Output: object

Output full type

{
}[] | {
  results?: {
  }[],
  data?: {
  }[],
  list?: {
  }[],
  totalResults?: number,
  nextPageToken?: string
} | {
}

Overview

Plugins

Competitions

downloadFile

downloadFiles

downloadLeaderboard

generateSubmissionUrl

list

listFiles

submit

viewLeaderboard

Datasets

create

createVersion

download

downloadFile

getMetadata

getStatus

list

listFiles

Kernels

downloadOutput

getStatus

list

listOutputFiles

pull

Models

get

getInstance

list

listInstanceVersionFiles

​Competitions

​downloadFile

​downloadFiles

​downloadLeaderboard

​generateSubmissionUrl

​list

​listFiles

​submit

​viewLeaderboard

​Datasets

​create

​createVersion

​download

​downloadFile

​getMetadata

​getStatus

​list

​listFiles

​Kernels

​downloadOutput

​getStatus

​list

​listOutputFiles

​pull

​Models

​get

​getInstance

​list

​listInstanceVersionFiles

Competitions

downloadFile

downloadFiles

downloadLeaderboard

generateSubmissionUrl

list

listFiles

submit

viewLeaderboard

Datasets

create

createVersion

download

downloadFile

getMetadata

getStatus

list

listFiles

Kernels

downloadOutput

getStatus

list

listOutputFiles

pull

Models

get

getInstance

list

listInstanceVersionFiles