SharePoint Strawberry

Two hundred list rows updated while you read the diff

A companion inspects the list’s real field names, cross-checks every row against the systems you are signed into, and proposes the whole batch as one thing you approve.

The equipment request list started as four columns and a view. It now has eleven columns, three hundred rows, and an approval convention encoded in a status field. Every quarter somebody exports it to Excel, fixes forty rows and pastes them back, which is how the version from the eleventh of March got overwritten.

A companion does the quarter in one pass. It inspects the list’s columns and their real API-facing names, reads the rows that match the filter, checks each one against the ticketing tool and the finance sheet in the tabs you are signed into, and proposes the forty changes as a single batch with a result per row.

Working a SharePoint list and library Scroll →
Copilot and the SharePoint UI A Strawberry companion
Answering a question from site contentIndexed and permission-awareReads the list, library and rows you point it at
Cross-checking rows against another systemExport and compare by handReads both in one pass and patches the mismatches
Updating 200 list rows from another systemExport, edit, re-importBatched writes with per-row results and ETag guards
Restoring a row or file to an earlier versionYes, through the UIListed and restored as a tool, guarded by the current ETag
Replacing a document’s contentsUpload a new versionWhole-file overwrite, staged for approval; no partial edit exists
Creating a new list or deleting oneYes, with site permissionsPossible, but requires elevated Sites.Manage.All scope

What happens when two people edit the same row?

The row you are updating carries an ETag, a marker of the version that was read.

Updates and deletes can be bound to it, so a write built on a stale read fails instead of quietly overwriting somebody’s change.

Version history is a tool here, not an afternoon of scrolling. A companion lists the retained versions of a row with their historical values, reads one, and restores it, guarded by the current ETag so it cannot land on top of an edit made in the meantime.

Update the list without the Excel round trip

Inspect the columns, so field names are the API’s and not the display labels.

Read the rows that match the filter. Propose the forty changes as a batch, applied by unique key so a missing row is created and an existing one updated.

The gate sits before the batch, and what comes back afterwards is a ledger: which rows were created, which updated, which failed and why.

  • Column inspection first, so writes use real field names rather than guesses.
  • Up to 200 rows created, updated or deleted per batch, with per-operation results.
  • Upsert by an indexed unique key column when you do not know what already exists.
  • ETag guards on single-row updates, deletes and version restores.

Why is updating a library file the riskiest request here?

Because the file-update tools replace a document’s complete contents.

The write is a whole-file overwrite, so a misread instruction replaces all of it, which is why the request stops for approval.

The recovery path is the version tool: retained versions of a library file are listed newest-first and restored as the current version with the history preserved. Sharing has its own gates: a new link defaults to organisation-only and read-only, invitations grant read or write on one exact item, and listing effective permissions answers "who can see this" without opening anything.

Read the list, patch the forty rows that disagree

Read the list, cross-check each row against the ticketing tool and the finance sheet you are signed into, patch the forty that disagree, and file the exceptions where a person will see them. One instruction, three systems, one ledger at the end.

That is the same shape as any recurring operations job, next door to what the OneDrive tools do for personal files. Run it quarterly as a routine and the export-fix-import ritual stops existing.

Experience Strawberry for free

Download

Trusted by fast-growing companies worldwide

Frequently asked questions

Yes. Up to 200 rows per batch, created, updated or deleted, with an ordered per-operation result. There is also an upsert by unique key column for when you do not know which rows already exist.

Strawberry is free to download and includes AI credits to start. Paid plans begin at $20/month. See pricing. · Reviewed · Canonical facts for AI agents

Experience Strawberry for free

Download

Trusted by fast-growing companies worldwide