The Kodexa CLI provides powerful commands for working with documents in document stores. This page explains how to query, upload, and manage documents.

​
Querying Documents

The query command allows you to search for and manipulate documents in a document store: 
kodexa query <store-ref> [<query>]
For example, to search for all documents in a store: 
kodexa query org-name/store-name "*"

​
Filtering Results

You can filter results using specific queries: 
kodexa query org-name/store-name "content:invoice"
Or use filters for more complex queries: 
kodexa query org-name/store-name "created>2023-01-01" --filter

​
Sorting Results

Sort results by specific fields: 
kodexa query org-name/store-name "*" --sort "name:asc"

kodexa query org-name/store-name "*" --sort "modified:desc"

​
Pagination and Streaming

For large result sets, use pagination: 
kodexa query org-name/store-name "*" --page 2 --pageSize 20
For continuous processing, use streaming: 
kodexa query org-name/store-name "*" --stream
Limit the number of results when streaming: 
kodexa query org-name/store-name "*" --stream --limit 100

​
Multithreaded Operations

Speed up operations with multiple threads: 
kodexa query org-name/store-name "*" --stream --threads 10

​
Downloading Documents

​
Download KDDB Format

Download documents in Kodexa Document Database (KDDB) format: 
kodexa query org-name/store-name "*" --download
This saves files as <document-path>.kddb

​
Download Native Files

Download the original native files: 
kodexa query org-name/store-name "*" --download-native
This saves files as <document-path>.native

​
Download Extracted Data

Download extracted data as JSON: 
kodexa query org-name/store-name "*" --download-extracted-data
This saves files as <document-path>-extracted_data.json Specify a project ID for extracted data: 
kodexa query org-name/store-name "*" --download-extracted-data --project-id project123

​
Reprocessing Documents

Reprocess documents using a specific assistant: 
kodexa query org-name/store-name "*" --stream --reprocess assistant-id
Note: Reprocessing requires streaming mode.

​
Labeling Documents

Add labels to documents: 
kodexa query org-name/store-name "*" --stream --add-label invoice
Remove labels: 
kodexa query org-name/store-name "*" --stream --remove-label draft

​
Watching for Changes

Monitor and refresh results periodically: 
kodexa query org-name/store-name "*" --watch 30
This refreshes the results every 30 seconds.

​
Deleting Documents

Delete documents matching a query: 
kodexa query org-name/store-name "*" --stream --delete
You’ll be prompted to confirm the deletion.

​
Uploading Documents

The upload command allows you to upload files to a document store: 
kodexa upload <store-ref> <file-paths>
For example, to upload a single file: 
kodexa upload org-name/store-name /path/to/document.pdf
To upload multiple files: 
kodexa upload org-name/store-name /path/to/document1.pdf /path/to/document2.pdf

​
Upload With External Data

You can attach external data to uploaded documents: 
kodexa upload org-name/store-name /path/to/document.pdf --external-data
This looks for a matching JSON file (e.g., document.json for document.pdf) and attaches its content as external data to the uploaded document.

​
Multithreaded Uploads

For faster uploads of multiple files: 
kodexa upload org-name/store-name /path/to/documents/*.pdf --threads 10

​
Advanced Examples

​
Search and download with multiple options

kodexa query org-name/store-name "path:*.pdf" --stream --download --download-extracted-data --threads 5

​
Reprocess all invoices with a new assistant

kodexa query org-name/store-name "labels:invoice" --stream --reprocess assistant-123

​
Monitor document store for changes

kodexa query org-name/store-name "*" --watch 60 --page 1 --pageSize 50

​
Complex filtering with sorting

kodexa query org-name/store-name "created>2023-01-01 AND modified<2024-01-01" --filter --sort "modified:desc"

​
Bulk operations with confirmation

# Add label to recent documents
kodexa query org-name/store-name "created>2024-01-01" --stream --add-label processed --threads 10

# Delete old drafts
kodexa query org-name/store-name "labels:draft AND modified<2023-01-01" --stream --delete

​
Upload with parallel processing

# Upload all PDFs from a directory with external metadata
kodexa upload org-name/store-name /data/documents/*.pdf --external-data --threads 10

​
Query Options Reference

OptionDescriptionExample
--filterUse filter syntax instead of query syntax--filter
--pagePage number for pagination--page 2
--pageSizeNumber of items per page--pageSize 20
--sortSort results by field--sort "name:asc"
--streamStream results instead of paginating--stream
--limitLimit number of results in streaming--limit 100
--threadsNumber of threads for operations--threads 10
--downloadDownload documents in KDDB format--download
--download-nativeDownload original files--download-native
--download-extracted-dataDownload extracted data as JSON--download-extracted-data
--project-idProject ID for extracted data--project-id proj123
--reprocessReprocess with assistant ID--reprocess asst123
--add-labelAdd label to documents--add-label invoice
--remove-labelRemove label from documents--remove-label draft
--deleteDelete matching documents--delete
--watchRefresh results every n seconds--watch 30

​
Upload Options Reference

OptionDescriptionExample
--threadsNumber of threads for parallel uploads--threads 10
--external-dataAttach JSON metadata from matching files--external-data