How does the bookkeeping and month-end-close example handle internal controls?+
The example models segregation of duties: the AI agent prepares and reconciles entries, and posting material or non-standard journal entries waits for a controller's approval at a human gate. Every entry carries its support and a tamper-evident audit hash. Which entries count as material, and who may approve them, is your control design to set. Minctrl holds no certification, and this is not legal advice.
How does the AI handle reconciliations and accruals?+
The AI agent ingests transactions and statements, categorizes and reconciles them, and drafts the period-end accruals and journal entries needed to tie the books to source records — surfacing anything it can't reconcile cleanly for a human rather than forcing a plug.
How does the example let an auditor trace a journal entry back to its source?+
The example models the trail: each entry is linked to the transactions and documents it came from, the person who approved it and an audit hash on a deterministic run, so the close can be reconstructed entry by entry. What your auditor accepts as evidence, and how long you keep it, is for you and your auditor to decide. Minctrl holds no certification, and this is not legal advice.