Sync schema-flexible data into a graph that doesn\'t need a fixed schema either.

VaultDB stores documents that don’t always share the same shape, which makes it hard to report on with traditional tools. Strand’s graph doesn’t require a fixed schema either, so nested and inconsistent documents map cleanly without flattening or losing structure.

How it works

Provide your VaultDB connection string and select the collections to sync. Cortex ingests documents in real time and resolves shared identifiers across collections, even when individual documents have different fields. Nested objects and arrays are preserved as structured relationships in the graph rather than collapsed into flat rows.

Configure

Manage the VaultDB connection from the integrations panel, where collection scope and connection credentials are set per workspace. Index hints can be provided to improve ingestion performance on large collections. Documents that don’t match expected identifiers are flagged for manual review rather than silently dropped.