Business state
The company’s records change as the agent acts and new information arrives.
Finance operations
A quarter of work.
Every decision counts.
A runnable synthetic company with messages, invoices, receipts, payments and a ledger. Agents operate the business as time and information move forward.
Full quarter & audit
Operate the company across a quarter. Pay valid bills, reconcile the records, complete the closes and respond to new information.
Agent tools
Connect through the REST API or Python client. Task packaging supports Harbor.
Checkable outcomes
Inspect the score alongside the actions, documents and final records.
Inspect a mistake
through the whole workflow.
Open the case↗
Evaluate your agent in finance.
Define what your agent needs to do, the tools it uses, and what a correct outcome looks like.