Knowledge: content import
Overview
The import brings existing collections into a knowledge space: an Obsidian vault, a folder of Markdown files or the export of an old wiki. The aim is that existing knowledge does not have to be retyped.
The import creates articles — it does not park them alongside. What is imported then behaves like a hand-written article: with versions, release and search.
Core tasks
Upload a collection. You choose the files and the target space. The folder structure is taken over as the outline.
Check the preview. Before applying you see what would be created: how many articles, which titles, which file could not be processed.
Apply. Only this step writes. Until then nothing has been created.
Fields in detail
The What it does column answers what changes in the system — not what the field is called. The grey name behind the label is the API field: the same thing runs under that name through automation, import and AI tools.
Starting an import
| Field | Required | Values / format | What it does |
|---|---|---|---|
File fileUlid | yes | an uploaded archive | The export from the source system. The structure of the archive becomes the structure of the articles; images are carried over. |
Target space spaceUlid | yes | an existing knowledge space | Where it is imported to. The space then governs access to everything imported. |
Article status articleStatus | no | published or draft | What the imported articles start on. Draft is the cautious choice: foreign content is reviewed first instead of appearing in search and AI answers straight away. |
Dry run dryRun | no | yes/no | Shows what would be created without creating anything. Before a large import, the move that saves an hour of cleaning up. |
Settings & permissions
The import requires knowledge.article.create — the same permission as creating by hand. In addition the server applies the space scope: you can only import into spaces you may contribute to.
Module flag: module.knowledge.
FAQ & troubleshooting
A file is missing from the result. The preview names the files that could not be processed — usually an unknown format or an empty file.
Images are missing. References to local images are only taken over if the images were uploaded too.
Titles are duplicated. The import creates, it does not merge. Duplicates can be merged in the space afterwards.