Each module is named for the standard it implements and keeps the record a reviewer needs: inputs, versions, citations and sign-off.
Record client acceptance and continuance decisions, collect independence declarations, and track identified threats and safeguards before planning begins. Firm-level ISQM 1 quality objectives and responses are managed alongside.
Risk areas carry inherent, control and detection risk against the governing standard, with a risk graph linking accounts, assertions and anomalies so the team can see where evidence needs to go.
Overall materiality is computed per engagement with performance materiality and the clearly-trivial threshold derived from it. Every revision is kept as a numbered version, so the file shows what changed and when.
Open the materiality fileJournal entries are scored for anomalies with an explanation per score. Computer-assisted audit techniques, exception lists and Benford first-digit analysis sit in the same workspace as the procedures they support.
Evidence files are scanned, hashed and attached to workpapers. External confirmations are tracked to response. Monetary unit sampling computes sample size from confidence factors, records item results and projects misstatement.
Model base and downside cash-flow scenarios from revenue, expenses, growth rates and opening cash. Test covenants against thresholds month by month and track key risk indicators, with an agent drafting the assessment for review.
Open the going concern fileSubsequent events are logged against the period end with their evaluation. Review notes, comments and tasks keep open points visible until they are cleared.
Draft audit reports by section, including key audit matters, and prepare engagement, representation and management letters. Reports are signed electronically, and financial statements can be tagged and validated for XBRL and ESEF export.
Each file shows how a capability behaves, with sample working papers and the limits we know about.
Agent definitions, prompt templates and the standards corpus are managed centrally. What reaches a workpaper has passed the checks below, or says plainly that it has not.
A question becomes a structured plan, the plan is validated against an allow-listed set of tables and fields, and the query runs under your tenant's row-level security. The model never writes the SQL.
Queryable areas include engagements, workpapers, findings, risk areas, agents and their runs, evidence, monitoring events, anomaly scores and the audit trail. Sessions and queries are logged.
Continuous monitoring rules evaluate on a schedule and record each trigger, so exceptions surface between fieldwork visits.
Connectors are built on the Model Context Protocol. Credentials are encrypted at rest. Most connectors are read-only; where write-back exists, each job needs partner and tenant-administrator approval. Clients without a connector upload through a request portal limited to their own document list.
The connector fileAvailability of individual connectors depends on your plan.
A demo covers your audit methodology, the standards you report under, and how agents, citations and review gates would fit your team. Pricing is discussed on the call.