> ## Documentation Index > Fetch the complete documentation index at: https://docs.twine.se/llms.txt > Use this file to discover all available pages before exploring further. # Get File Transfers > List file transfers. These records are kept indefinitely, including the identifier each target system assigned, even though the uploaded files themselves are removed shortly after delivery. ## OpenAPI ````yaml https://api.twine.se/spec/openapi.json get /v1/org/file-transfers openapi: 3.0.3 info: contact: email: help@twine.se name: Twine Support description: >+ # Introduction Welcome to the Twine Public API reference. This API allows you to read data from Twine's system, and to write data back for supported domains. See the Pushing data section below for what can be written. # Versioning The API is versioned. The current version is v1.1, served under the `/v1` path. As the API is subject to change, new versions may be released. Always refer to the latest documentation to ensure you're using the correct version. ## Dated model attributes In v1.1, every entity field is tracked by change date. This allows you to fetch data as it existed at a specific point in time, or retrieve all changes from a specific point in time up to the present. You can also supply a shape to the API to get data in a format that suits your needs. ### Dated values In the v1.1 model, all fields are represented by an array of values with an optional valid_from timestamp. A null value for valid_from indicates that the field value is valid from the beginning of time, meaning it has no recorded history in the system. When multiple values exist in the array, they are sorted in descending order with the most recent value first. The last value may or may not have a null timestamp. An array can never contain more than one value with a null timestamp. # Authentication The Twine public API uses Bearer token authentication. The token should be sent in the `Authorization` header with the value `Bearer `. ## Provisioning a token To provision a token, you must first generate a long-lived refresh token using Twine backoffice. This token can then be used to generate short-lived access tokens. The refresh token should be kept secret and never shared. Follow these steps to obtain an access token: 1. Invoke `/provisioning/refresh-token/begin` using the JTI you received when provisioning the refresh token. 2. The response will contain a `salt` value 3. Concatenate the salt with the refresh token, separated by a colon (`:`) 4. Hash the concatenated string using SHA256, e.g. `sha256 ":"` 5. The hash should be in lower case 6. Invoke `/provisioning/refresh-token/refresh` with the JTI (as `jti`) and the hash (as `signature`) in the body 7. The response will contain an access token ## Legacy tokens Legacy tokens are still supported. These tokens are long-lived and can be used to authenticate against the API. However, we recommend using the new token provisioning system. ## Token expiration Provisioned tokens are short-lived. Decode them to find the expiry time. Legacy tokens have longer expiry times. # Filtering Some endpoints allow filtering on certain fields. Not all fields are filterable, and filter operations may vary between fields. Refer to specific endpoints to discover what is possible. All filters follow the same structure: ```bash GET /v1/org/employees?filters[updated_at][value]=2022-01-01T00:00:00Z&filters[updated_at][op]=gt ``` # Ordering Some endpoints allow ordering on certain fields. Not all fields are orderable. Refer to specific endpoints to discover what is possible. All ordering follows the same structure: ```bash GET /v1/org/employees?order[field]=updated_at&order[direction]=desc ``` # Data reliability A given piece of data is a snapshot of the latest data fetched from the Vendor. In some cases the data returned by Twine will be out of sync with the Vendor. This is due to the nature of the integration and the data provided by the Vendor. Twine will always strive to provide the most accurate data possible. Depending on the Customer, the data may be updated in near real-time, or with a longer delay. # Permissions All API requests are subject to a permissions check. Permissions are configured by Twine staff or an approved external admin. # Rate limiting The API is not currently rate limited. However, you should expect this to change in the near future. At that time you will start receiving 429 Too Many Requests responses. More information will be provided here when rate limiting is implemented. ## Fetching single resources When requesting single resources with inadequate permissions, you will receive a 403 Forbidden in response. ## Fetching lists of resources When requesting a list of resources, the API will filter out resources you do not have access to. This means that the response may contain fewer resources than requested. The API will always return a 200 OK response, even if the list is empty. # Pushing data NOTE: The API currently only supports the `employee` domain for writing data. There are multiple considerations to take into account when writing data to the API. * In order to write data, the role being used must have the `create` and/or `update` permissions for the `employee` domain. * As per the RESTful API design specs, the `POST` method is used to create new resources, while the `PUT` method is used to update existing resources, and the `PATCH` method is used to partially update existing resources. * Even though `PATCH` will not touch unspecified dated properties, it will still replace all entries of targeted dated properties with the new values provided in the request body. This means that if you want to keep existing historical values for a given dated property, you must include them in the request body. This behaviour can be changed in the future given enough demand. * If there are incoming domain mappings for the `employee` domain, you should avoid updating any properties that are mapped in those integrations. Otherwise there will be uncertainty about which integration is responsible for the data. * If there is a configured `twine` integration with outgoing domain mappings for the `employee` domain, a replication job will be created to push the changes to the configured integration. If you want to avoid this, create a condition in the target integration. # Nomenclature Throughout the API, we use the following terms: | Term | Description | |-------------|-------------| | Customer | The entity on whose behalf Twine processes personnel data | | Vendor | Third party supplier of data with which Twine integrates on behalf of Customer | # FAQ ### Do you have a sandbox environment? Yes. Please contact Twine support to get access. ### Are there webhooks? Not yet, but it is on the roadmap. ### How can I keep the response body small? Use the `select` parameter to only fetch the fields you need. This will reduce the response size and speed up the request. Combine this with filtering by `updated_at` to only fetch the data that has changed since your last request. title: | Twine Public API version: | v1.2 x-logo: altText: Twine Logo backgroundColor: '#00000000' url: >- https://images.squarespace-cdn.com/content/v1/63cec8699b6d396be083ed8b/a2746b2d-2582-49f0-93bb-99a8765c5450/LogoTwineHorizontal.png?format=12w servers: - description: Twine Public Production API url: https://api.twine.se variables: {} - description: Twine Public Staging API url: https://api-stage.twine.se variables: {} security: - BearerAuth: [] tags: - description: >+ # Fetching Employees can be retrieved in paginated lists, or one by one. The employee object remains the same in both cases. # Data consistency Twine allows Customers to map their data to their own wishes. In some cases, fields might not contain what you expect. Fields like `first_name` and `private_id_number` are more or less guaranteed to contain what you expect. However, fields like `phone1`, `phone2` might mean different things depending on the Customer. Some Customers will want to store an employee's private phone number in `phone1`, while others will store the work phone number instead. Always refer to the Customer's data mapping (upcoming API endpoint) to understand what the fields contain. name: Employees - description: > # Fetching Organizational units can be retrieved in paginated lists, or one by one. The organizational unit object remains the same in both cases. # Data consistency Twine will attempt to normalize the source system's organizational unit tier. If successful, the `tier` field will contain the normalized tier and be of type [OrganizationalUnitTier](#model/organizationalunittier) name: Organizational Units - description: >+ # Fetching Time reports can be retrieved in paginated lists. # Time types Twine allows Customers to map their time types to their own wishes. In some cases, time types might not contain what you expect. name: Time Reports paths: /v1/org/file-transfers: get: tags: - File Transfers summary: Get File Transfers description: >- List file transfers. These records are kept indefinitely, including the identifier each target system assigned, even though the uploaded files themselves are removed shortly after delivery. operationId: getFileTransfers parameters: - description: '' in: query name: pagination required: false schema: description: Cursor pagination properties: after: description: >- Cursor to start after. Is returned in the response. Must be specified together with `first`. Leave undefined to start at the beginning nullable: true type: string before: description: >- Cursor to start before. Is returned in the response. Must be specified together with `last`. Leave undefined to start at the end nullable: true type: string first: description: >- Number of subsequent items to return. Must be specified together with `after`. Will be capped at 100, unless otherwise specified nullable: true type: integer last: description: >- Number of previous items to return. Must be specified together with `before`. Will be capped at 100, unless otherwise specified nullable: true type: integer title: CursorPagination type: object - description: '' in: query name: order required: false schema: $ref: '#/components/schemas/FileTransferOrder' - description: Filter parameters for File Transfer in: query name: filters required: false schema: $ref: '#/components/schemas/FileTransferFilter' responses: '200': content: application/json: schema: $ref: '#/components/schemas/FileTransferListResponse' description: Response '400': content: application/json: schema: $ref: '#/components/schemas/BadRequestError' description: Bad Request Response '403': content: application/json: schema: $ref: '#/components/schemas/ForbiddenError' description: Forbidden Response '500': content: application/json: schema: $ref: '#/components/schemas/ForbiddenError' description: Internal Error Response callbacks: {} components: schemas: FileTransferOrder: description: Parameters for File Transfer ordering properties: direction: enum: - asc - desc type: string field: enum: - inserted_at - updated_at - file_name type: string required: - field - direction title: FileTransferOrder type: object FileTransferFilter: description: Parameters for File Transfer filtering properties: employee_id: description: Twine employee UUID the document belongs to properties: op: enum: - eq - in type: string value: type: string required: - value - op type: object file_name: description: Uploaded file name properties: op: enum: - eq - in type: string value: type: string required: - value - op type: object id: description: File transfer UUID properties: op: enum: - eq - in type: string value: type: string required: - value - op type: object inserted_at: description: Inserted at timestamp properties: op: enum: - eq - gt - lt - gte - lte type: string value: format: date-time type: string required: - value - op type: object system_integration_id: description: Target system integration UUID properties: op: enum: - eq - in type: string value: type: string required: - value - op type: object updated_at: description: Updated at timestamp properties: op: enum: - eq - gt - lt - gte - lte type: string value: format: date-time type: string required: - value - op type: object title: FileTransferFilter type: object FileTransferListResponse: description: A paginated list of file transfers properties: data: items: $ref: '#/components/schemas/FileTransfer' type: array pagination: $ref: '#/components/schemas/CursorPaginationResult' required: - data - pagination title: FileTransferListResponse type: object BadRequestError: description: Bad request error properties: code: example: 400 type: integer errors: default: [] items: description: Singular error properties: domain: type: string message: type: string reason: type: string required: - domain - reason - message title: InnerError type: object type: array message: example: Bad Request type: string required: - code - message - errors title: BadRequestError type: object ForbiddenError: description: Forbidden error properties: code: example: 403 type: integer errors: default: [] items: description: Singular error properties: domain: type: string message: type: string reason: type: string required: - domain - reason - message title: InnerError type: object type: array message: example: Forbidden type: string required: - code - message - errors title: ForbiddenError type: object FileTransfer: description: >- A record of one file Twine delivered to a target system. Twine does not keep uploaded files — they are removed shortly after delivery — but this record is kept, including the identifier the target system assigned to the document. properties: bytes_retain_until: description: >- When the uploaded file itself is removed. This record and the identifiers on it are kept indefinitely; the file is not. format: date-time nullable: true type: string content_sha256: description: >- SHA-256 of the uploaded file, hex encoded. Useful for verifying what was sent. nullable: true type: string content_type: example: application/pdf nullable: true type: string delivered_at: description: When the target accepted the file. format: date-time nullable: true type: string delivery_job_id: description: The job that performed the upload to the target, once one has run. format: uuid nullable: true type: string domain_mapping_id: description: The domain mapping used to deliver the file. format: uuid nullable: true type: string employee_id: description: >- The Twine employee the document belongs to, if any. Twine resolves this to the employee's id in the target system when delivering. format: uuid nullable: true type: string file_name: description: The uploaded file's name example: first-aid-2026.pdf type: string id: description: File transfer UUID format: uuid type: string idempotency_key: nullable: true type: string inserted_at: format: date-time type: string job_id: description: >- The job that carried out the transfer. Look it up under `/v1/org/jobs` for detail. format: uuid nullable: true type: string metadata: additionalProperties: true description: The metadata supplied on upload, returned unchanged. type: object org_id: description: Organization UUID format: uuid type: string remote_employee_id: description: >- The employee's id in the target system, when the document hangs off an employee there. Some targets only address a document as the pair of this and `remote_id`. example: E-10422 nullable: true type: string remote_id: description: >- The identifier the target system assigned to the uploaded document. Null until the transfer is delivered, and also null afterwards for targets that do not return one (file area targets identify a document by path, and some HR systems return nothing at all on upload) — or when a document of that name was already present and the domain mapping does not overwrite, in which case nothing was uploaded and `remote_metadata.note` says so. example: 3fa85f64-5717-4562-b3fc-2c963f66afa6 nullable: true type: string remote_metadata: additionalProperties: true description: >- Any further identifying detail the target returned. May also carry a `note` explaining an outcome the other fields cannot express — notably a transfer that is `delivered` with no `remote_id` because a document of that name was already present and the domain mapping does not overwrite, so nothing was uploaded and the target was never asked for an identifier. type: object remote_path: description: >- Where the file landed in the target, for targets addressed by path rather than by id (SFTP, S3). example: /certificates/2026/first-aid-2026.pdf nullable: true type: string size: description: Size in bytes of the uploaded file nullable: true type: integer status: description: > Where the transfer has got to. - `pending` — accepted and on its way to the target. Poll until this changes. - `delivered` — the target accepted the file. `remote_id` and `remote_path` hold whatever identifier it returned. - `rejected` — the file reached the target's domain mapping but was deliberately not sent: a condition on the system integration or domain mapping excluded it, or that mapping is disabled. Nothing malfunctioned, so there is no error to look up. - `failed` — delivery was attempted and did not succeed. Look up `job_id` under `/v1/org/jobs` for the reason. - `undeliverable` — no file transfer domain mapping from Twine matched, so nothing was sent. The file was recorded but never delivered. - `unknown` — the underlying job record is no longer available. This record outlives the job, so a very old transfer can end up here. Derived from the delivery job rather than stored, so it cannot be filtered on. enum: - pending - delivered - rejected - failed - undeliverable - unknown example: delivered type: string system_integration_id: description: The target system integration the file was delivered to. format: uuid nullable: true type: string updated_at: format: date-time type: string required: - id - org_id - file_name - status title: FileTransfer type: object CursorPaginationResult: description: Cursor pagination result properties: end_cursor: description: Cursor to start after to retrieve the next page nullable: true type: string has_next_page: description: If true, there are additional items available after the current page type: boolean has_previous_page: description: >- If true, there are additional items available before the current page type: boolean start_cursor: description: Cursor to start before to retrieve the previous page nullable: true type: string title: CursorPaginationResult type: object securitySchemes: BearerAuth: bearerFormat: JWT description: Bearer token for authentication scheme: bearer type: http ````