View, Edit, and Export ML-BOM
View, edit, and export ML-BOM for models in JFrog AI Catalog Registry.
Use the ML-BOM tab in JFrog AI Catalog Registry to review model metadata, complete fields on custom models, and export a machine-readable bill of materials. For field definitions and editability rules, see Understanding ML-BOM in AI Catalog.
This guide covers the following tasks:
View an ML-BOM
The ML-BOM tab shows a two-column table of fields and values, a search bar to filter fields, and a Last modified timestamp.
To view the ML-BOM for a model in AI Catalog:
-
From the JFrog Platform menu, select AI/ML > Registry.
-
In the project tree, select the model.
-
Open the ML-BOM tab using the path for your model type:
-
External models and model packages: On the model page, select ML-BOM (tabs: Model info, Runtime, ML-BOM).
-
Custom models (version scope): Open the model version (for example, NeuroScan-Insight > Version 1.0.4), then select ML-BOM (tabs: Overview, Builds, ML-BOM).
-
Custom models (build scope): From Builds, open a build, then select ML-BOM (tabs: Logs, Code, Security, Deployments, ML-BOM).
-
-
Optionally enter text in the search bar to filter fields in the table.
-
Review the Last modified timestamp on the ML-BOM tab.
Users with Read Model permission can view ML-BOM records.
Edit ML-BOM Fields
Editing requires Manage Model permission. Which fields you can edit depends on model type. See Field Editability by Model Type.
To edit ML-BOM fields:
-
Open the ML-BOM tab for the model, version, or build (see View an ML-BOM).
-
Click Edit.
-
Enter or update editable fields.
-
Click Save to persist your changes, or Cancel editing to discard them.
Export ML-BOM
Export produces a JSON file in CycloneDX v1.7 or SPDX v3.0 format.
To export an ML-BOM:
-
Open the ML-BOM tab for the model, version, or build (see View an ML-BOM).
-
Click Export ML-BOM.
-
In the Export ML-BOM dialog, select a format:
- CycloneDX v1.7 (JSON)
- SPDX v3.0 (JSON)
-
Click Export. The file downloads to your browser.
Frequently Asked Questions
This section provides answers to frequently asked questions about viewing, editing, and exporting ML-BOM in AI Catalog.
FAQs
Q: Which export formats does ML-BOM support?
A: The Export ML-BOM dialog supports CycloneDX v1.7 (JSON) and SPDX v3.0 (JSON).
Q: Can I edit ML-BOM on an external model?
A: External and package models harvest most fields as read-only. You can edit Risk Assessment when Edit is available on the ML-BOM tab. Other fields reflect provider or catalog metadata at harvest time.
Q: Where are field definitions documented?
A: See the 13-field reference and editability table in Understanding ML-BOM in AI Catalog.
