How MARCReady Handles Your Catalogue Data
How MARCReady processes uploaded catalogue files, what data is retained, and how optional anonymised contribution works.
MARCReady is designed for bibliographic catalogue records. This article explains what happens when you upload a file, what data is retained, and what you should not upload.
This is a practical summary. For formal wording, see the Privacy and GDPR page and Terms.
What MARCReady is designed to process
MARCReady is designed for bibliographic catalogue records including MARC21 files, MARCXML files, MRK files, and CSV, TSV, Excel, and JSON catalogue exports.
MARCReady is not designed for patron or borrower records.
What not to upload
Do not upload files containing: patron names, patron addresses, patron email addresses, borrower records, circulation history, fines or payment history, staff account exports, passwords or credentials, or sensitive personal notes unrelated to bibliographic description.
If such data is uploaded by mistake, contact [email protected] so KohaSupport can help arrange deletion.
What happens when you upload a file
When you upload a file to MARCReady:
- The file is stored temporarily for processing.
- MARCReady analyses the file format and structure.
- Kai may assist with field mapping and repair suggestions.
- You review the preview and suggested changes.
- MARCReady creates output files where export is available.
- Uploaded source files and processed outputs are deleted according to the published retention schedule.
Standard retention
| Data | Purpose | Retention |
|---|---|---|
| Uploaded source file | Processing and review | Deleted within 2 days after processing is complete |
| Preview and processed output | Review and export download | Up to 90 days |
| File hash | Duplicate handling and deterministic preview selection | Same as processed output |
| Temporary bibliographic cache | Avoid redundant processing | Up to 90 days |
Optional anonymised contribution
MARCReady includes an optional opt-in setting called “Contribute anonymised MARC correction examples”. This setting is disabled by default.
If you enable it, KohaSupport may use selected anonymised bibliographic correction examples to improve MARCReady’s repair quality. You can use MARCReady without enabling this option. Contribution can be disabled at any time.
This optional contribution should not include patron data, borrower records, circulation history, payment data, staff accounts, passwords, or uploaded source files as complete files.
Use of Amazon Bedrock
MARCReady may use Amazon Bedrock for AI-assisted MARC field mapping and repair suggestions. Amazon Bedrock is an AWS managed AI service. AWS states that Bedrock does not use customer prompts or outputs to train Amazon or third-party foundation models.
KohaSupport’s optional MARCReady improvement contribution is separate from Amazon Bedrock’s foundation model training. If you choose to contribute anonymised correction examples, those examples are used only to improve MARCReady.
Payment processing
MARCReady paid plans are billed through Stripe. KohaSupport does not store payment card details directly.
Related resources
Next Steps
More in Koha System
Was this article helpful?