Skip to content

Data

← Admin console

The data viewer, at /data. It lets you browse the records OpenSteward has synced, with personal data masked by default.

The data viewer shows what the data store actually holds for each connection, without database access and without exposing personal data by accident. Use it to:

  • check that a sync brought back what you expected;
  • answer a support question;
  • export a slice of records.
  • data:read (operator, admin) opens the viewer. Records are always masked.
  • data:export (operator, admin) downloads CSV or NDJSON.
  • data:read_pii (admin) reveals one record’s personal data. It also allows Include personal data on an export.
  • A tenant-restricted user sees only their tenants’ connections. Another tenant’s connection answers “not found”.
  • See what’s stored. The index lists every connection in your scope with:

    • its label and full connection id, with a Copy button;
    • its vertical, tenant and platform;
    • one badge per resource type, with a record count;
    • the total number of records;
    • when the data last changed. Hover the time to see the exact date.

    If the deployment runs in passthrough mode, nothing is stored and the page says so.

  • Find a connection. The index works like the other admin tables:

    • The row under the headers filters by Label, Connection id, Vertical, Tenant, Platform and Resource types. Text boxes match part of what the column shows, ignoring case. An id prefix is enough, and typing invoices under Resource types keeps only connections that hold invoices. Press Enter or Search to apply, and Clear to reset.
    • Click any column header to sort by it. Click again to reverse the order. Records sorts by count, and Last change by the time behind “17 h ago”.
    • Filters and sort order live in the URL, so a filtered view can be bookmarked or shared.
  • Browse a connection.

    • Each resource type has its own tab. The table is cursor-paginated.
    • Search the resource, filter on the fields that allow it, sort by last updated or record id, and pick columns with Columns.
    • Column headers carry PII, Financial and Secret badges.
  • Open a record. Click its id to open it in a dialog. You see:

    • its connection, platform, fetch and update times and schema version;
    • the record as a JSON tree with classification badges;
    • a field-classification table showing how each field is masked.
  • Reveal personal data (data:read_pii).

    • Give an optional reason and press Reveal personal data.
    • The unmasked record is shown once and the reveal is written to the audit log, with your name and the reason.
    • Reload the page to go back to the masked view.
  • Export (data:export). Download the current resource as CSV or NDJSON.

    • The export is masked the same way as the table.
    • Holders of data:read_pii can tick Include personal data.
    • Every export is audited.
  • Personal-data fields are masked, and secrets are removed.
  • Financial fields are masked too when an admin turns on data_access.mask_financial in Settings.
  • A resource type with no field classification yet is masked by default, and the page warns you about it.

The data index: record counts and freshness per connection

A connection’s records: resource tabs, search, filters, columns and export

A single record, masked, with its JSON tree and classification badges