core candidate

OfficeOps-200

General desktop workflows across email, calendar, files, spreadsheets, PDFs, forms, web research, downloads, uploads, and multi-app handoffs.

Core release candidate: OfficeOps-50

Task families

20 families x 10 variants

EmailDraftEmailAttachmentSaveCalendarCreateCalendarRescheduleSpreadsheetUpdateFileRenameFileMovePDFExtractFormFillWebResearchToSheetDownloadAndUploadFolderCreateDocFormatSlideUpdateContactCreateReminderCreateSearchAndSummarizeCSVTransformScreenshotCompareMultiAppWorkflow

Target UIs

Mock apps for resettable rollouts

Gmail-like mailbox

Calendar mock

Drive/File Explorer mock

Sheets/Excel-like editor

PDF workspace

Examples

Representative task contracts

Example 1

OfficeOps.EmailAttachmentSave.001

Instruction: Find the latest email from [email protected] and save the attached invoice PDF to Invoices/AWS.
Grader: file exists in Invoices/AWS and source email/message id matches latest AWS billing email

Example 2

OfficeOps.SpreadsheetUpdate.001

Instruction: Open June Leads.xlsx and mark all rows with status Contacted as Follow-up needed.
Grader: all Contacted rows updated and unrelated rows unchanged

Example 3

OfficeOps.DownloadAndUpload.001

Instruction: Download the invoice from the supplier portal and upload it to the accounting system.
Grader: downloaded invoice hash matches uploaded attachment hash on correct accounting record

Verifier

Suite-level verifier shape

Filesystem state, document diff, spreadsheet cell checks, calendar/email records, and artifact presence.