The previous nested folder upload flow bypassed the backend batch-upload
contract when parentId was set. Instead of creating the whole metadata
tree in one backend operation, the frontend recursively called
createFolder/getFileUploadUrl for each node.
That introduced two regressions for uploads into subfolders:
- consistency regression: mid-sequence failures could leave partially
created folder trees under the destination folder
- performance regression: metadata creation degraded from a single
batch request to O(files + folders) round-trips before file bytes
were uploaded
This change moves nested uploads back to the original batch semantics:
- add optional parent_id support to app asset batch-upload payload
- create the whole nested tree under the target parent in
AppAssetService.batch_create_from_tree
- pass parentId through useBatchUpload instead of using per-node
createFolder/getFileUploadUrl calls
- remove the now-unnecessary useBatchUploadOperation wrapper
- add a backend unit test covering batch tree creation under an
existing parent folder
After this change, both root uploads and subfolder uploads use the same
single-request metadata creation path, preserving atomic tree creation
semantics and avoiding avoidable metadata round-trips.
Wire up the "Create Blank Skill" action card to open a modal where
users enter a skill name. The modal validates against existing skill
names in real-time and creates a folder with a SKILL.md file via
batchUpload, then opens the file as a pinned tab.
Change onSuccess to onSettled for upload mutations so the file tree
refreshes regardless of success or failure, ensuring consistency when
backend creates nodes but storage upload fails.
Previously, media files were fetched via getFileContent API which decodes
binary data as UTF-8, resulting in corrupted strings that cannot be used
as img/video src. Now media files use getFileDownloadUrl API to get a
presigned URL, enabling proper preview of images and videos of any size.
When file content is not in JSON format (e.g., newly uploaded files),
return the raw content instead of empty string to ensure files display
correctly.
Extract repeated appId retrieval and tree data fetching patterns into
dedicated hooks (useSkillAssetTreeData, useSkillAssetNodeMap) to reduce
code duplication across 6 components and leverage TanStack Query's
select option for efficient nodeMap computation.
