Overview
This node integrates with the Dust API to perform two main operations: interacting with an AI assistant ("Talk to an Agent") and uploading documents to a specified space within the Dust platform. The "Upload a Document" operation allows users to send textual content along with metadata to be stored and processed in Dust's document management system.
Typical use cases for the "Upload a Document" operation include:
- Automating the ingestion of textual data into Dust workspaces for further analysis or AI processing.
- Adding documents programmatically from other systems or workflows, tagging them, and associating metadata such as MIME type or source URL.
- Managing document uploads asynchronously to avoid blocking workflow execution.
For example, a user might upload product manuals or customer support transcripts into a Dust space to enable AI-powered search or summarization.
Properties
| Name | Meaning |
|---|---|
| Space ID | The identifier of the Dust space where the document will be uploaded. |
| Data Source Name | The name of the data source within the space that will contain the document. |
| Document ID | A unique identifier for the document being uploaded. |
| Document Content | The full text content of the document to upload. |
| Additional Fields | A collection of optional fields to provide extra metadata or control upload behavior: |
| - Title | The title of the document. |
| - MIME Type | The MIME type of the document (e.g., "text/plain", "application/pdf"). |
| - Source URL | A URL pointing to the original source of the document. |
| - Tags | Comma-separated tags to associate with the document for categorization or filtering. |
| - Async Upload | Boolean flag indicating whether the upload should be performed asynchronously (true/false). |
| - Light Document Output | Boolean flag indicating whether to return a lightweight version of the document response, excluding large fields like text, chunks, and vectors. |
Output
The node outputs a JSON array containing the response(s) from the Dust API after attempting to upload the document(s). The structure of each item corresponds to the Dust API's document upload response, which typically includes details about the uploaded document such as its ID, status, and metadata.
If the "Light Document Output" option is enabled, the output excludes heavy fields like the full text content, document chunks, and vector embeddings, providing a more concise summary of the document.
No binary data output is produced by this operation.
Dependencies
- Requires an active connection to the Dust API via an API key credential.
- The node uses the workspace ID and region information from the credentials to construct API endpoints.
- Network access to Dust's API endpoints (
https://dust.ttorhttps://eu.dust.ttdepending on region) is necessary. - Proper configuration of the API key and workspace ID in n8n credentials is required.
Troubleshooting
Common Issues:
- Invalid or missing API credentials will cause authentication failures.
- Incorrect Space ID, Data Source Name, or Document ID may result in 404 Not Found errors.
- Providing malformed or empty document content can lead to API validation errors.
- Network connectivity issues to Dust API endpoints can cause request timeouts or failures.
Error Messages:
- Errors returned from the Dust API are captured and surfaced by the node. They typically include HTTP status codes and error messages.
- If the node is configured to continue on failure, errors are included in the output JSON with details; otherwise, the node execution stops with an error.
- Common error resolution steps include verifying all required parameters, checking API key validity, and ensuring the target space and data source exist.
Links and References
- Dust API Documentation (for detailed API usage and document upload specifications)
- n8n Documentation on Creating Custom Nodes
- General info on MIME Types