Open finance skill
Vendor rule audit
Are automated categorization rules still supported by transaction history?
Compare vendor rules with real historical outcomes and surface stale, conflicting, ambiguous, or overly broad logic.
Inputs
- ✓Vendor rules
- ✓Categorized history
- ✓Manual overrides
Outputs
- ✓Stable rules
- ✓Conflict queue
- ✓Proposed narrow changes
Checks performed
- ✓Historical consistency
- ✓Override frequency
- ✓Rule scope
Guided synthetic example
See the check from input to decision
One broad vendor rule categorized transactions consistently last year but conflicts with recent purchases.
This is the operating model we use at Level. The percentages are estimated shares of the work in a typical review. Actual effort varies with data quality, complexity, and the issue found.
Start with the source records
About 15% of work| Vendor | Rule account | Transactions | Overrides |
|---|---|---|---|
| Example Marketplace | Office supplies | 84 | 19 |
| Sample Fuel | Vehicle fuel | 46 | 1 |
Run the deterministic check
About 20% of workMeasure rule coverage and override frequency, then inspect whether conflicts reflect vendor ambiguity or a stale rule.
Surface the flagged result
About 10% of workNeeds review
Example Marketplace has a 23% override rate, too high for an unconditional office-supplies rule.
Use recorded finance judgment to analyze possible reasons
About 25% of workAt Level, we record the review logic, known explanations, and questions our finance professionals use in this situation. AI applies that documented human judgment to organize the most plausible reasons, without pretending it knows which reason is true.
Possible reasons to investigate
The vendor sells multiple categories, so a narrower memo, amount, cardholder, or class condition may work better than deleting automation.
Make the final human judgment
About 20% of workA Level finance professional validates the source evidence, challenges the AI-assisted analysis, and decides which explanation is supported.
Make the operating decision
About 10% of workStage a narrower rule proposal and have a finance professional test it against historical exceptions before approval.
Use this system
Start with the example, then inspect the structure
- Download the invented input and expected output.
- Replace one field at a time with a safe test value.
- Compare the result with the expected structure.
- Have a finance professional review every exception before acting.
MCP preview
run_synthetic_example {"slug":"vendor-rule-audit"}Human review is part of the system
This check organizes evidence and surfaces exceptions. A person with finance domain expertise still needs to interpret the result, validate the source records, and discuss the operating decision.
Open source does not mean open client data. We publish reusable finance tools, checks, templates, and playbooks. Client records, workpapers, credentials, communications, and private configurations remain confidential.