Bank Reconciliation Agent
Read bank statements, assign payments, post bank fees, clarify differences.
Reads bank statements in CAMT.053 and MT940 format, assigns payments via exact and fuzzy matching, posts bank fees and interest automatically and escalates unassignable items.
Score Dashboard
What This Agent Does
Bank reconciliation is a daily obligation in financial accounting. Bank statements must be read in, every item assigned to a posting, bank fees posted and differences clarified. In a company with multiple bank accounts and hundreds of daily transactions, this ties up a clerk for hours.
The Decision Layer breaks bank reconciliation into seven decision steps. Bank statements in CAMT.053 or MT940 format are automatically read in. Exact assignments (amount plus reference) run rule-based. Fuzzy assignments (partial amounts, deviating references) use AI pattern matching. Bank fees and interest are automatically posted. Only unassignable items are escalated to the clerk.
The result: daily bank reconciliation is done in minutes instead of hours. The reconciliation report is created automatically. And continuous reconciliation ensures no surprises wait at month-end.
Micro-Decision Table
Read bank statement Are bank statement data correctly parsed? Rules Engine
Parsing per CAMT.053 or MT940 standard
Decision Record
Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.
Exact assignment Can the payment be uniquely assigned to an open item? Rules Engine
Database match on amount and reference
Decision Record
Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.
Fuzzy assignment Can the payment be assigned via pattern matching? AI Agent Vendor
Pattern matching for partial amounts and deviating references
Decision Record
Challengeable: Yes - fully documented, reviewable by humans, objection via formal process.
Challengeable by: Vendor
Post bank fees To which account are bank fees posted? Rules Engine
Mapping table by transaction type
Decision Record
Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.
Post interest To which account is interest posted? Rules Engine
Mapping table by interest type (credit/debit)
Decision Record
Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.
Escalate unassignable items Must a clerk manually clarify the item? Human Vendor
Clarification requires human judgement
Decision Record
Challengeable: Yes - via manager, works council, or formal objection process.
Challengeable by: Vendor
Create reconciliation report Is the report created and archived? Rules Engine
Formatted report per template
Decision Record
Challengeable: Yes - rule application verifiable. Objection possible for incorrect data or wrong rule version.
Decision Record and Right to Challenge
Every decision this agent makes or prepares is documented in a complete decision record. Affected parties (employees, suppliers, auditors) can review, understand, and challenge every individual decision.
Prerequisites
- Bank interface (HBCI/FinTS, EBICS) or statement import (CAMT.053, MT940)
- ERP system with open items (receivables, payables)
- Mapping table for bank fees and interest
- Configured confidence threshold for fuzzy assignments
Governance Notes
GoBD-relevant: bank transactions are tax-relevant and must be recorded promptly (cashless transactions within 10 days per GoBD). Bank statements are posting documents subject to the retention obligation per AO Paragraph 147 (8 years per BEG IV).
The reconciliation report is part of the procedural documentation. The statutory auditor reviews bank reconciliation as a standard audit procedure. Complete assignment reports accelerate the financial audit.
§203 StGB-relevant data is encrypted end-to-end and never passed to AI models in plain text.
Process Documentation Contribution
Infrastructure Contribution
The Bank Reconciliation Agent builds the banking interface infrastructure (CAMT.053, MT940, EBICS) reused by the Payment Traffic Agent and Cash Forecasting Agent. The pattern matching framework for fuzzy assignments is used by the Reconciliation Agent and Cash Application Agent. Daily bank reconciliation provides the data basis for continuous reconciliation.
Builds Decision Logging and Audit Trail used by the Decision Layer for traceability and challengeability of every decision.
Does this agent fit your process?
We analyse your specific finance process and show how this agent fits into your system landscape. 30 minutes, no preparation needed.
Analyse your processRelated Agents
Payment Traffic Agent
Format payments, transmit to bank, process acknowledgements, ensure four-eyes review.
Cash Forecasting Agent
Create liquidity forecast - recognise historical patterns, model scenarios, highlight action needs.
Frequently Asked Questions
Does the agent work with all German banks?
Yes. The agent processes the standard formats CAMT.053 and MT940 supported by all German banks. Connection is via HBCI/FinTS or EBICS - depending on the bank and transaction volume.
What happens with bank statements in foreign currency?
Foreign currency transactions are converted at the daily rate and matched against the posted EUR amount. Exchange rate differences are automatically recognised and posted as gains/losses.
How fast is the agent compared to manual reconciliation?
With 100+ daily transactions, reconciliation time drops from 2-3 hours to 15-20 minutes. The clerk only handles the 5-10% of items that cannot be automatically assigned.
What Happens Next?
30 minutes
Initial call
We analyse your process and identify the optimal starting point.
1 week
Discover
Mapping your decision logic. Rule sets documented, Decision Layer designed.
3-4 weeks
Build
Production agent in your infrastructure. Governance, audit trail, cert-ready from day 1.
12-18 months
Self-sufficient
Full access to source code, prompts and rule versions. No vendor lock-in.
Implement This Agent?
We assess your finance process landscape and show how this agent fits your infrastructure.