Unofficial SDK — generated from @codeforge-main/checkr-sdk's public OpenAPI specification by the CodeForge platform. Not affiliated with or endorsed by @codeforge-main/checkr-sdk.

codeforge-main-checkr-sdk

Version 1.3.0

Full generator behavior documentation: https://www.code-forge.net/docs

Requires Node 20+ (or a modern browser) — the generated transport uses AbortSignal.any and AbortSignal.timeout.

Install

npm install codeforge-main-checkr-sdk

yarn add codeforge-main-checkr-sdk

pnpm add codeforge-main-checkr-sdk

Usage

This SDK is exposed as a single, fully typed client class. Construct it once with your shared configuration and call any API operation as a method — each returns the deserialized response alongside HTTP metadata.

import { CheckrComApiDocsClient } from 'codeforge-main-checkr-sdk'

const client = new CheckrComApiDocsClient({
	baseUrl: 'https://api.checkr.com/v1',
	auth: { type: 'basic', username: 'YOUR_API_USERNAME', password: 'YOUR_API_PASSWORD' },
})

const response = await client.account()

console.log(response.status) // HTTP status code
console.log(response.data)   // typed as AccountResponse

The client holds the shared configuration (baseUrl, auth, retry, hooks). Each method accepts a per-call context extending HttpClientContext — supply the operation-specific payload / parameters, and override any shared field per call.

Available operations

• account
• createAccount
• getAdverseAction
• cancelAdverseAction
• getCandidateStory
• deleteCandidateStory
• getFormI9
• createFormI9
• getFormI9Id
• getFormI9Pdf
• listOfCandidates
• createCandidate
• getCandidateDocuments
• uploadDocument
• listOfDriverLicenses
• createDriverLicense
• getDriverLicense
• updateDriverLicense
• deleteDriverLicense
• employersListByCandidate
• createEmployer
• getEmployer
• employerDelete
• schoolsListByCandidate
• createSchoolForCandidate
• listOfProfessionalLicenses
• createProfessionalLicense
• deleteProfessionalLicenses
• getSchool
• schoolDelete
• getCandidateSsn
• getCandidatePostalAddress
• createPostalAddress
• getCandidate
• updateCandidate
• deleteCandidate
• getCounties
• getCountyCriminalSearch
• getDocument
• getDrugAlcoholClearinghouseSearches
• getEducationVerification
• getEmploymentVerification
• getPersonalReferenceVerification
• getProfessionalReferenceVerification
• getFederalCriminalSearch
• getFederalDistrictCriminalSearch
• getFederalCivilSearch
• getFederalDistrictCivilSearch
• geosList
• createGeo
• getGeo
• updateGeo
• deleteGeo
• getGlobalWatchlistSearch
• getAccountHierarchy
• updateAccountHierarchy
• createAccountHierarchyNodes
• getAccountHierarchyStatus
• listOfInvitations
• createInvitation
• getInvitation
• cancelInvitation
• getInternationalAdverseMediaSearch
• getInternationalCriminalSearch
• getInternationalEducationVerificaiton
• getInternationalEmploymentVerification
• getInternationalGlobalWatchlistSearch
• getInternationalIdentityDocumentValidation
• getInternationalMotorVehicleReport
• getMotorVehicleReport
• getNationalCriminalSearch
• nodesList
• getNode
• updateNode
• deleteNode
• packagesList
• getPackage
• deletePackage
• getProfessionalLicenseVerification
• programsList
• getProgram
• createAdverseAction
• createReport
• getReport
• updateReport
• completeReport
• requestReportEta
• getTagsForReport
• addTagToReport
• updateTagsForReport
• deleteTagFromReport
• addressesListByReport
• adverseItemsByReport
• assessmentsByReport
• createCandidateStory
• verificationsListByReport
• getSexOffenderSearch
• getSsnTrace
• getStateCriminalSearch
• getSubscriptions
• createSubscription
• getSubscription
• updateSubscription
• cancelSubscription
• listOfContinuousChecks
• createContinuousCheck
• getContinuousCheck
• postContinuousCheck
• cancelContinuousCheck
• usersList
• getVerification
• webhooksList
• createWebhook
• getWebhook
• deleteWebhook
• worksitesList
• worksitesCreate
• getWorksites
• updateWorksites
• deleteWorksites
• getFacisSearch

Transport

Every request goes through the same transport: a per-attempt timeout, cancellation via AbortSignal, and an optional custom fetch.

const client = new CheckrComApiDocsClient({
	timeoutMs: 10_000, // default 60000ms, applied fresh to each retry attempt
})

const controller = new AbortController()
setTimeout(() => controller.abort(), 5_000)

await client.account({ signal: controller.signal })

timeoutMs is per attempt, not per call — a request that times out and then retries gets a fresh timer on each attempt, and throws TimeoutError if every attempt times out. A caller-triggered AbortSignal is never retried; a timeout is, under the same rule as a network error. Streaming operations ignore timeoutMs and support only caller cancellation.

Custom fetch

Pass a fetch implementation to route every request through it — a proxy included, via undici's ProxyAgent:

import { ProxyAgent, fetch as undiciFetch } from 'undici'

const dispatcher = new ProxyAgent('http://localhost:8080')
const client = new CheckrComApiDocsClient({
	fetch: (url, init) => undiciFetch(url, { ...init, dispatcher }),
})

hooks.makeRequest wins when both are set — it replaces the whole transport, and fetch is read only by the built-in one.

Raw requests

An endpoint the generated operations don’t model is still reachable, through the same auth, retries, timeout and hooks:

const { data, response } = await client.get('/v1/status')
await client.post('/v1/events', { body: { type: 'ping' } })

// the general form, for a method the verb helpers don’t cover:
await client.request({ method: 'OPTIONS', path: '/v1/status' })

path is joined onto baseUrl with the URL constructor — a leading / resolves from the origin (dropping any path segment baseUrl itself carries); a relative segment appends after it. A plain-object body is JSON-serialized with a JSON Content-Type; pass an already-encoded value (a string, FormData, ...) to send anything else. data parses by the response’s own Content-Type: JSON via .json(), a textual type via .text(), anything else as a Blob.

Uploads

A multipart operation's binary part accepts a File directly. The toFile helper builds one from bytes, a Blob, or a Node stream — fs.createReadStream(...) included, since every Node stream is async-iterable. The source is read fully into memory before the request goes out, so this is not a fit for uploads too large to hold in memory at once.

import { toFile } from 'codeforge-main-checkr-sdk'
import fs from 'node:fs'

const file = await toFile(fs.createReadStream('./report.pdf'), 'report.pdf', {
	contentType: 'application/pdf',
})

// pass `file` as the operation's multipart field, e.g. `{ payload: { file } }`

A Blob, a Uint8Array (a Node Buffer included), or a File you already have works the same way, without toFile.

Errors

A non-OK response throws one of these HttpError subclasses, picked by an exact status match first, then any >= 500 response as InternalServerError, then plain HttpError for anything else. Every one of them is also an instanceof HttpError, so a catch-all keeps working alongside a narrower check.

| Class | Status | | --- | --- | | BadRequestError | 400 | | AuthenticationError | 401 | | PermissionDeniedError | 403 | | NotFoundError | 404 | | ConflictError | 409 | | UnprocessableEntityError | 422 | | RateLimitError | 429 | | InternalServerError | any >= 500 |

import { HttpError, NotFoundError } from 'codeforge-main-checkr-sdk'

try {
	await client.account()
} catch (error) {
	if (error instanceof NotFoundError) {
		// error.status is narrowed to 404
	} else if (error instanceof HttpError) {
		// any other non-OK response — error.status is number
	}
	throw error
}

Validation

Every model is plain data — the shape it has on the wire. This SDK carries no schemas and no runtime dependency, so nothing is validated for you: response.data is the parsed body typed as the operation declares it.

Regenerate the SDK with validation schemas turned on to get a zod schema exported beside every model type.

Generated by CodeForge v0.86.9.