InFlow
Skill pack

Docs

Backup & restore

A full backup is a ZIP you save: inflow-db.json plus the attachment files currently on this device. Import restores locally first. Drive sync is a separate choice.

What is in the ZIP

inflow-db.json
attachments/
  {mediaFileId or fileName}

Nested layouts such as InFlow/inflow-db.json are also accepted. A standalone JSON (no attachments) still imports vault data only.

Export

Profile → export writes the vault JSON plus every attachment that is available on the device (durable local path or media cache). Media that exists only on Drive and is not cached is omitted from that ZIP. If you need those files, open the entries so they cache, or copy them from Drive.

Import

  1. 1Pick the ZIP (or a standalone inflow-db.json).
  2. 2InFlow validates required collections (folders, views or legacy tabs, entries, tags) and migrates the schema.
  3. 3Nothing is written until you confirm. Import always saves locally first.
  4. 4Drive sync — new vault or existing — is a separate step. Matched attachments are re-uploaded so this app owns new file ids under drive.file.

A failed Drive upload leaves the local restore recoverable.

Trash is not backup

Soft delete hides entries, workspaces, and views (deletedAt). Restore from Profile → Trash. There is no permanent purge button in this version. Uninstalling the app or clearing storage removes the local cache. Drive copies stay until you delete the folder.

Using a backup with the skill pack

You can attach inflow-db.json from a backup to an AI tool with the skill pack. That is a copy you chose to share with that model. Merge any proposed JSON back yourself — the skill does not write to Drive.