> ## Documentation Index
> Fetch the complete documentation index at: https://composio-preview.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Cradl AI

> Cradl AI is an API-first platform for building and operating AI-powered document extraction workflows.

![Cradl AI logo](https://mintcdn.com/composio-preview/NPf8agM6UrvZD-Iu/images/toolkits/cradl_ai.svg?fit=max\&auto=format\&n=NPf8agM6UrvZD-Iu\&q=85\&s=69906060a058b53aa8191e67b4394fcc)

* **Category:** documents & ocr
* **Auth:** S2S\_OAUTH2
* **Composio-managed OAuth available?** No
* **Tools:** 11
* **Triggers:** 0
* **Slug:** `CRADL_AI`
* **Version:** 20261101\_00

## Tools

> [accordion] Download File
>
> **Slug:** `CRADL_AI_DOWNLOAD_FILE`
>
> Download document, OCR, annotation, prediction, or run-variable content from an exact Cradl AI file-service URL with a cradl:file ID.
>
> #### Input Parameters
>
> **file\_url** `string` *required*
>
> Exact [https://files.api.cradl.ai](https://files.api.cradl.ai) URL returned by a Cradl AI resource.
>
> **output\_name** `string`
>
> Optional filename for the downloaded file.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Agent Run
>
> **Slug:** `CRADL_AI_GET_AGENT_RUN`
>
> Get the current state and outputs of one Cradl AI agent run. Poll this read-only tool until the run reaches a terminal status; it does not itself consume page quota.
>
> #### Input Parameters
>
> **run\_id** `string` *required*
>
> Bare run identifier returned by Start Agent Run or List Agent Runs, such as cradl:run:... .
>
> **agent\_id** `string` *required*
>
> Agent identifier that owns the run.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Document
>
> **Slug:** `CRADL_AI_GET_DOCUMENT`
>
> Get metadata, processing links, and annotations for one Cradl AI document by ID. This read does not consume page quota.
>
> #### Input Parameters
>
> **document\_id** `string` *required*
>
> Document identifier returned by List Documents or document creation.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Prediction
>
> **Slug:** `CRADL_AI_GET_PREDICTION`
>
> Get one extracted prediction, including its status, fields, errors, and downloadable output references. This read does not consume page quota.
>
> #### Input Parameters
>
> **prediction\_id** `string` *required*
>
> Prediction identifier returned by List Predictions or a run output.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Agent Runs
>
> **Slug:** `CRADL_AI_LIST_AGENT_RUNS`
>
> List processing runs for an agent, optionally filtered by status or creation/update time. Use this to discover recent extraction jobs and their run IDs.
>
> #### Input Parameters
>
> **status** `string ("archived" | "exported" | "pending-export" | "pending-predictions" | "ready-for-review" | "review-completed" | "succeeded-predictions" | "error" | "running" | "completed" | "review-in-progress")`
>
> Return only runs in the selected provider-defined status.
>
> **agent\_id** `string` *required*
>
> Agent identifier whose runs should be listed.
>
> **max\_results** `integer`
>
> Maximum runs to return in this page; Cradl does not document an upper bound.
>
> **next\_cursor** `string`
>
> Opaque continuation token returned by a previous call. Omit for the first page and pass it back without modification.
>
> **updated\_time\_after** `string`
>
> Return only runs with updatedTime after this ISO 8601 timestamp.
>
> **created\_time\_before** `string`
>
> Return only runs with createdTime before this ISO 8601 timestamp.
>
> **updated\_time\_before** `string`
>
> Return only runs with updatedTime before this ISO 8601 timestamp.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Agents
>
> **Slug:** `CRADL_AI_LIST_AGENTS`
>
> List document-processing agents available to the connected Cradl AI workspace. Use an agent ID from this result to start an extraction run.
>
> #### Input Parameters
>
> **max\_results** `integer`
>
> Maximum agents to return in this page; Cradl does not document an upper bound.
>
> **next\_cursor** `string`
>
> Opaque continuation token returned by a previous call. Pass it back unchanged, or omit it for the first page.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Documents
>
> **Slug:** `CRADL_AI_LIST_DOCUMENTS`
>
> List documents in the connected Cradl AI workspace and discover document IDs and content/output file URLs.
>
> #### Input Parameters
>
> **order** `string ("ascending" | "descending")`
>
> Sort documents in ascending or descending order.
>
> **sort\_by** `string`
>
> Sort documents by their creation timestamp.
>
> **consent\_id** `string`
>
> Optional consent-ID filter.
>
> **dataset\_id** `string`
>
> Optional dataset-ID filter.
>
> **document\_id** `string`
>
> Optional exact document-ID filter.
>
> **max\_results** `integer`
>
> Maximum documents to return in this page; Cradl does not document an upper bound.
>
> **next\_cursor** `string`
>
> Opaque continuation token returned by a previous call. Pass it back unchanged, or omit it for the first page.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] List Predictions
>
> **Slug:** `CRADL_AI_LIST_PREDICTIONS`
>
> List extracted predictions, optionally filtered by model, and discover prediction IDs and output file URLs.
>
> #### Input Parameters
>
> **order** `string ("ascending" | "descending")`
>
> Sort predictions in ascending or descending order.
>
> **sort\_by** `string`
>
> Sort predictions by their creation timestamp.
>
> **model\_id** `string`
>
> Optional model identifier to filter predictions.
>
> **max\_results** `integer`
>
> Maximum predictions to return in this page; Cradl does not document an upper bound.
>
> **next\_cursor** `string`
>
> Opaque continuation token returned by a previous call. Omit it for the first page and pass it back unchanged to continue.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Process Document
>
> **Slug:** `CRADL_AI_PROCESS_DOCUMENT`
>
> Create a Cradl agent run and document, then upload one document for asynchronous extraction, consuming processed-page quota. The tool returns after upload without polling; poll the run using the original agent ID and returned bare run ID, and use the composite agent run ID for document attachment. On certain failures it attempts to delete the created document and run; do not automatically retry an ambiguous upload outcome.
>
> #### Input Parameters
>
> **file** `object` *required*
>
> One PDF, image, or email file to process.
>
> **name** `string`
>
> Optional persistent document name.
>
> **agent\_id** `string` *required*
>
> Exact Cradl agent ID that should process the document.
>
> **metadata** `object`
>
> Optional caller metadata to persist on the document.
>
> **variables** `object`
>
> Optional variables made available to the new agent run.
>
> **description** `string`
>
> Optional persistent document description.
>
> **content\_type** `string ("application/pdf" | "image/jpeg" | "image/png" | "image/tiff" | "image/webp" | "message/rfc822")` *required*
>
> MIME type to record for the uploaded document.
>
> **retention\_in\_days** `integer`
>
> Optional document retention period from 1 through 1825 days.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Start Agent Run
>
> **Slug:** `CRADL_AI_START_AGENT_RUN`
>
> Create a processing run for an existing Cradl AI agent. This creates persistent run state; attaching and processing document content against the run consumes the workspace's monthly processed-page quota.
>
> #### Input Parameters
>
> **agent\_id** `string` *required*
>
> Agent identifier from List Agents.
>
> **variables** `object`
>
> Optional variables made available to the agent run.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Upload Document to Run
>
> **Slug:** `CRADL_AI_UPLOAD_DOCUMENT_TO_RUN`
>
> Create a document attached to a Cradl AI agent run and upload its PDF, image, or email content for asynchronous processing, consuming processed-page quota. On certain failures, the tool attempts to delete the created document.
>
> #### Input Parameters
>
> **file** `object` *required*
>
> PDF, image, or email file to upload.
>
> **name** `string`
>
> Optional human-readable document name.
>
> **metadata** `object`
>
> Optional caller metadata to store on the document.
>
> **description** `string`
>
> Optional document description.
>
> **agent\_run\_id** `string` *required*
>
> Composite agent run ID from Start Agent Run's id field, containing the agent and run IDs and optionally an organization prefix.
>
> **content\_type** `string ("application/pdf" | "image/jpeg" | "image/png" | "image/tiff" | "image/webp" | "message/rfc822")` *required*
>
> MIME type of the document file.
>
> **retention\_in\_days** `integer`
>
> Optional retention period in days, from 1 through 1825.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not
