| name | da-live |
| description | Working with Adobe Document Authoring (da.live) content mounted as a VFS filesystem. Use when reading or writing files through a DA mount (da:// source). Non-sheet files use normal file operations; sheet files (.json) must go through the da-live script which handles schema validation, derived fields, and a single atomic write.
|
| allowed_tools | ["bash","read_file","write_file","edit_file"] |
| scripts | ["da-live.jsh"] |
Adobe Document Authoring (DA) — VFS Mount
DA content is accessed via a filesystem mount backed by https://admin.da.live.
Files are read and written as plain bytes — DA performs no server-side processing of content.
Sheet files (.json)
Sheets are a special case. Always use the da-live script to write them — it validates
rows against the existing schema, recalculates derived fields, and performs a single
write to the remote. Never write sheet JSON directly to the VFS path.
Non-sheet files
For all other file types (HTML, markdown, images, etc.), use normal file operations
directly on the VFS path — read, edit, and write as you would any local file.
Working with sheets
-
Inspect the schema before constructing rows:
da-live schema /mnt/<mount-path>/data/sheet.json
-
Write the updated row set — validates schema, recalculates, and writes in one operation:
da-live write /mnt/<mount-path>/data/sheet.json '[{"key":"val",...}, ...]'
da-live write /mnt/<mount-path>/data/sheet.json --sheet <name> '[{"key":"val",...}, ...]'
-
Verify by re-reading the file:
cat /mnt/<mount-path>/data/sheet.json
Key rules
- Always use
da-live write to modify sheets — never write JSON directly to the VFS path.
- Always pass the complete row set for the target sheet, not just changed rows. For multi-sheet files, other sheets are preserved automatically.
- Rows must exactly match the existing schema. The script will fail with a clear error
if any row has missing or extra keys.
- For multi-sheet files, unmodified sheets are preserved as-is.
