Suites
DesktopBench workflow suites
Browse DesktopBench suites. Each suite targets 20 task families, 10 variants per family, resettable mock environments, executable verifiers, and optional human trajectories.
Last updated at
core candidate
StoreOps-200
20 task families.
Commerce operator workflows across product catalog, inventory, orders, refunds, promotions, shipping, buyer chat, and sales analytics.
core candidate
AccountingOps-200
20 task families.
Accounting workflows with naturally verifiable outcomes: invoices, receipts, expenses, bank reconciliation, tax codes, reports, exports, and month close.
core candidate
SupportOps-200
20 task families.
Support workflows across ticket classification, replies, macros, customer lookup, order lookup, escalations, follow-ups, and closure.
1K candidate
SalesOps-200
20 task families.
CRM workflows across leads, contacts, companies, deals, stages, notes, follow-ups, proposals, quotes, forecasts, and exports.
core candidate
OfficeOps-200
20 task families.
General desktop workflows across email, calendar, files, spreadsheets, PDFs, forms, web research, downloads, uploads, and multi-app handoffs.
expansion candidate
LegalOps-200
20 task families.
Legal desktop workflows across contract upload, clause tagging, signature requests, matter updates, risk flags, obligations, and renewal tracking.
expansion candidate
HROps-200
20 task families.
HR workflows across employee records, time off, payroll review, onboarding, offboarding, documents, candidates, equipment, and access.
KR differentiator
GovOps-200
20 task families.
Public portal workflows inspired by Korean government and tax portals: certificates, tax payments, applications, document upload, receipts, and status checks.