Skip to content
Changelog

Report a bug

Uploadcare integration: operations for AI agents

Uploadcare connects to Type.com as an integration. This page lists the 48 operations documented for Claude and Codex in Type.com: 22 read, 17 write, 9 destructive.

About Uploadcare

Uploadcare vendor API references, subject to account permissions.

Categories: Files & Storage

How to connect Uploadcare to Type.com

Access
Organization connection: one shared account for the workspace.

To add the connection, follow the steps in Space connections.

Documented operations

48 operations. Your connection's permissions determine which operations are available. API references describe the vendor surface; they do not expand your connection's credentials or access.

48 operations listed.

REST / File

  • Read

    List of files

    Getting a paginated list of files. If you need multiple results pages, use `previous`/`next` from the response to navigate back/forth.

    GET /files/

  • Read

    File info

    Get file information by its UUID (immutable).

    GET /files/{uuid}/

  • Read

    Search files

    Search files by filename, metadata, MIME type, UUID, or tags using full-text or exact matching, with optional filtering by size, upload date, and file type. Use `query` for a quick search across all full-text fields at once. Use `phrase` to search specific fields. Use `exact`, `datetime_uploaded`, `size`, `is_image`, or `tags` for structured filtering. At least one condition is required. **How conditions combine:** all top-level conditions are combined with AND — a file must satisfy every condition you pass. There is no top-level OR and no general NOT. OR is available only within a single `exact` field (multiple values), within `query` (across fields), and in `tags.any`; NOT is available only via `tags.none` and `is_image: false`. A field cannot appear in both `phrase` and `exact` in the same request. Full-text matching (`query`, `phrase`) is substring-aware: a term matches anywhere inside a word. Each value must be at least 4 characters; for short or precise values use `exact`. Set `fuzziness: true` to enable typo-tolerant search. Disabled by default. **Note:** Enabling fuzziness significantly increases query latency. **Note:** Newly uploaded files are indexed asynchronously and may not appear in search results immediately. See the [file search guide](https://uploadcare.com/docs/file-search/) for worked examples.

    POST /files/search/

  • Write

    Store file

    Store a single file by UUID. When file is stored, it is available permanently. If not stored — it will only be available for 24 hours. If the parameter is omitted, it checks the `Auto file storing` setting of your Uploadcare project identified by the `public_key` provided in the `auth-param`.

    PUT /files/{uuid}/storage/

  • Write

    Batch file storing

    Used to store multiple files in one go. Up to 100 files are supported per request. A JSON object holding your File list SHOULD be put into a request body.

    PUT /files/storage/

  • Write

    Copy file to local storage

    POST requests are used to copy original files or their modified versions to a default storage. Source files MAY either be stored or just uploaded and MUST NOT be deleted. Copying of large files is not supported at the moment. If the file CDN URL includes transformation operators, its size MUST NOT exceed 100 MB. If not, the size MUST NOT exceed 5 GB. Note: In the immediate 201 response, the `result.size` can be `0` and the `result.mime_type` can be `application/octet-stream`. This indicates the copy has been initiated but is not yet completed. Once the copy finishes, fetching the file info (e.g., `GET /files/{uuid}/`) will return the actual size and MIME type.

    POST /files/local_copy/

  • Write

    Copy file to remote storage

    POST requests are used to copy original files or their modified versions to the S3 bucket. Source files MAY either be stored or just uploaded and MUST NOT be deleted. Copying of large files is not supported at the moment. File size MUST NOT exceed 5 GB.

    POST /files/remote_copy/

  • Destructive

    Delete file

    Removes individual files. Returns file info. Note: this operation removes the file from storage but doesn't invalidate CDN cache.

    DELETE /files/{uuid}/storage/

  • Destructive

    Batch file delete

    Used to delete multiple files in one go. Up to 100 files are supported per request. A JSON object holding your File list SHOULD be put into a request body. Note: this operation removes files from storage but doesn't invalidate CDN cache.

    DELETE /files/storage/

REST / Add-Ons

  • Read

    Check AWS Rekognition execution status

    Check the status of an Add-On execution request that had been started using the [Execute Add-On](#operation/awsRekognitionExecute) operation.

    GET /addons/aws_rekognition_detect_labels/execute/status/

  • Read

    Check AWS Rekognition Moderation execution status

    Check the status of an Add-On execution request that had been started using the [Execute Add-On](#operation/awsRekognitionDetectModerationLabelsExecute) operation.

    GET /addons/aws_rekognition_detect_moderation_labels/execute/status/

  • Read

    Check ClamAV execution status

    Check the status of an Add-On execution request that had been started using the [Execute Add-On](#operation/ucClamavVirusScanExecute) operation.

    GET /addons/uc_clamav_virus_scan/execute/status/

  • Read

    Check Remove.bg execution status

    Check the status of an Add-On execution request that had been started using the [Execute Add-On](#operation/removeBgExecute) operation.

    GET /addons/remove_bg/execute/status/

  • Write

    Execute AWS Rekognition

    Execute [AWS Rekognition](https://docs.aws.amazon.com/rekognition/latest/dg/labels-detect-labels-image.html) Add-On for a given target to detect labels in an image. **Note:** Detected labels are stored in the file's appdata.

    POST /addons/aws_rekognition_detect_labels/execute/

  • Write

    Execute AWS Rekognition Moderation

    Execute [AWS Rekognition Moderation](https://docs.aws.amazon.com/rekognition/latest/dg/moderation.html) Add-On for a given target to detect moderation labels in an image. **Note:** Detected moderation labels are stored in the file's appdata.

    POST /addons/aws_rekognition_detect_moderation_labels/execute/

  • Write

    Execute ClamAV

    Execute [ClamAV](https://www.clamav.net/) virus checking Add-On for a given target.

    POST /addons/uc_clamav_virus_scan/execute/

  • Write

    Execute Remove.bg

    Execute [remove.bg](https://remove.bg/) background image removal Add-On for a given target.

    POST /addons/remove_bg/execute/

REST / File metadata

  • Read

    Get file's metadata

    Get file's metadata keys and values.

    GET /files/{uuid}/metadata/

  • Read

    Get metadata key's value

    Get the value of a single metadata key.

    GET /files/{uuid}/metadata/{key}/

  • Destructive

    Update metadata key's value

    Update the value of a single metadata key. If the key does not exist, it will be created.

    PUT /files/{uuid}/metadata/{key}/

  • Destructive

    Delete metadata key

    Delete a file's metadata key.

    DELETE /files/{uuid}/metadata/{key}/

REST / File tags

  • Read

    Get file tags

    Returns the current ordered list of tags for a file.

    GET /files/{uuid}/tags/

  • Destructive

    Replace file tags

    Replaces the entire tag set of a file. Tags not in the new list are deleted. An empty array clears all tags. Tags are normalized before storage (lowercased, whitespace-stripped, duplicates removed). Calling PUT with the same set of tags (regardless of case or order) produces no write — `added` and `deleted` will both be empty.

    PUT /files/{uuid}/tags/

  • Destructive

    Add or remove file tags

    Adds and/or removes tags in a single atomic operation. `add` and `delete` are both optional and default to `[]`. Delete is applied first, then add — a tag can be removed and re-added in the same request. Tags in `add` that are already present are silently skipped. Tags in `delete` that are not present are silently ignored. If the resulting tag count would exceed 50, the request fails with 400. An empty body `{}` is valid and returns the current tag state unchanged. Tags are normalized before storage (lowercased, whitespace-stripped, duplicates removed).

    PATCH /files/{uuid}/tags/

REST / Group

  • Read

    List of groups

    Get a paginated list of groups.

    GET /groups/

  • Read

    Group info

    Get a file group by its ID. Groups are identified in a way similar to individual files. A group ID consists of a UUID followed by a “~” (tilde) character and a group size: integer number of the files in the group.

    GET /groups/{uuid}/

  • Destructive

    Delete group

    Delete a file group by its ID. **Note**: The operation only removes the group object itself. **All the files that were part of the group are left as is.**

    DELETE /groups/{uuid}/

REST / Project

  • Read

    Project info

    Getting info about account project.

    GET /project/

REST / Webhook

  • Read

    List of webhooks

    List of project webhooks.

    GET /webhooks/

  • Write

    Create webhook

    Create and subscribe to a webhook. You can use webhooks to receive notifications about your uploads. For instance, once a file gets uploaded to your project, we can notify you by sending a message to a target URL.

    POST /webhooks/

  • Destructive

    Update webhook

    Update webhook attributes.

    PUT /webhooks/{id}/

  • Destructive

    Delete webhook

    Unsubscribe and delete a webhook.

    DELETE /webhooks/unsubscribe/

REST / Conversion

  • Read

    Document info

    The endpoint allows you to determine the document format and possible conversion formats.

    GET /convert/document/{uuid}/

  • Read

    Document conversion job status

    Once you get a conversion job result, you can acquire a conversion job status via token. Just put it in your request URL as `:token`.

    GET /convert/document/status/{token}/

  • Read

    Video conversion job status

    Once you get a processing job result, you can acquire a processing job status via token. Just put it in your request URL as `:token`.

    GET /convert/video/status/{token}/

  • Write

    Convert document

    Uploadcare allows you to convert files to different target formats. Check out the [conversion capabilities](https://uploadcare.com/docs/transformations/file-conversion/#document-file-formats) for each supported format.

    POST /convert/document/

  • Write

    Convert video

    Uploadcare video processing adjusts video quality, format (mp4, webm, ogg), and size, cuts it, and generates thumbnails. Processed video is instantly available over CDN.

    POST /convert/video/

UPLOAD / Upload

  • Read

    Check the status of a task to fetch/upload a file from a URL

    The endpoint notifies about the progress of the task by changing the value of the `status` field: * the field is set to `waiting` initially * once the system starts processing the request, the field is set to `progress` * if the file is fetched and uploaded successfully, the `status` field is set to `success` and the endpoint returns a JSON object holding information about the uploaded file * if the system fails to fetch/upload the file, the `status` field is set to `error` and the `error` field is set to the cause of the error.

    GET /from_url/status/

  • Read

    Get information about an uploaded file

    Returns a JSON object with information about an uploaded file (file size, MIME type, metadata, etc).

    GET /info/

  • Write

    Direct uploads

    Direct file uploads comply with the [RFC 7578](https://www.rfc-editor.org/rfc/rfc7578) standard. In other words, you can perform a direct upload by making an HTTP POST request with the Content-Type header set to `multipart/form-data`. **Note**: Direct file uploads support files smaller than 100 megabytes only. If you would like to upload larger files, please use [Multipart Uploads](/docs/api/upload/upload/multipart-file-upload-start) instead. File upload example with curl: ``` curl -X POST https://upload.uploadcare.com/base/ \ -F UPLOADCARE_PUB_KEY="demopublickey" \ -F UPLOADCARE_STORE="auto" \ -F "metadata[subsystem]"="uploader" \ -F "metadata[pet]"="cat" \ -F file=@"my file.jpeg" ```

    POST /base/

  • Write

    Start multipart upload

    Multipart uploads should be used if you need to upload files larger than 100 megabytes or if you want to explicitly trigger [AWS S3 Transfer Acceleration](https://aws.amazon.com/s3/transfer-acceleration/). When you use Multipart Uploads your files go straight to AWS S3 bypassing our upload instances. To upload large files, 3 requests are sent in sequence: 1. `/multipart/start/` 2. `<presigned-url-x>` 3. `/multipart/complete/`. **Note**: Multipart uploads support files larger than 10 megabytes only.

    POST /multipart/start/

  • Write

    Upload individual file parts

    The second phase is about uploading the file parts to the presigned upload URLs returned from the `/multipart/start/` endpoint. Each uploaded part should be at least 5 MiB (5242880 bytes) in size except for the last one, which can be smaller. You can upload the file parts in parallel provided that the byte order stays unchanged. **Note**: You MUST define `Content-Type` header for your data.

    PUT /<presigned-url-x>

  • Write

    Complete multipart upload

    Once all the file parts have been uploaded successfully, complete the upload session to assemble all the file parts into a single resulting file.

    POST /multipart/complete/

  • Write

    Upload files from URLs

    Uploadcare can fetch a file from a publicly available URL and then automatically upload the fetched file to your project. #### Upload tokens Requests to the endpoint return a JSON dictionary with a `token` that can be further used to [check the status](/docs/api/upload/upload/from-url-upload-status) of the upload request. **Note:** The token is not a file ID and can't be used to address the file directly. The actual file ID should be retrieved by calling the `/from_url/status/` endpoint. #### Duplicates prevention By default, every call to the `/from_url/` endpoint with the same `source_url` results in a new upload leading to file duplication. If you would like Uploadcare to keep track of the requested URLs and avoid the duplicate uploads, pass the `save_URL_duplicates` and `check_URL_duplicates` parameters described below.

    POST /from_url/

UPLOAD / Groups

  • Read

    Get information about a file group

    Returns a JSON object with information about a file group (when the group was created, number of the files in the group, etc).

    GET /group/info/

  • Write

    Create a file group

    Create a file group from a set of already uploaded files. The most common use case for creating a file group is when a user uploads multiple files at once and then wants to display them together. **Note:** A group itself and files within it MUST belong to the same project. **Note:** Groups are immutable and the only way to add/remove a file to a group is to create a new one.

    POST /group/

URL / File information

  • Read

    File info as JSON

    Returns file-related information, such as image dimensions or geo tagging data in the JSON format.

    GET /{uuid}/-/json/

URL / File names

  • Read

    File with a custom name

    You can set an optional filename that users will see instead of the original name.

    GET /{uuid}/{filename}

See How connections work for access and permissions, or browse all integrations.