Journal Entry Testing & Fraud Risk Screening
Scores every journal entry posted in a period against a fixed, documented list of fraud-risk factors (per AU-C 240 / SAS 99) before a person even looks at one — reads straight from QuickBooks, and refuses to score an incomplete or unverified batch of entries.
- Plugin
- audit-analytics
- Version
- v1.0.9
- Impact
- Very High
- Complexity
- Medium
How it works
- Input
- A period's journal entries pulled from QuickBooks (or a spreadsheet from any accounting system), the chart of accounts, and — if available — who prepared and approved each entry, plus the HR termination list
- What it does
- Checks the population is complete before doing anything else → converts every timestamp to one stated time zone → scores every entry against a documented list of risk factors → pulls a sample using a mix of directed and random selection → drafts a possible explanation for each risky cluster, then has a second, independent pass try to knock that explanation down
- What has to hold true
- Every flagged entry has to point to the specific number and threshold that tripped it — no flag based on a vague impression. Rules that depend on frequency (like unusual timing patterns) won't fire unless there's enough history to compare against.
- Output
- A 9-tab Excel workpaper — the scored population, the list of exceptions, the sample selected for testing, a separate list of anything that couldn't be tested for lack of information, and a built-in validation tab
The part that makes it reproducible
Building blocks it runs on
Shared, independently tested components. They ship bundled inside this skill — there is nothing extra to install.
What you'll need to change for your firm
This runs as published, but every audit function documents things its own way. These are the places it assumes a house format:
- Your accounting system and export format, if not QuickBooks
- Your chart of accounts and which accounts carry elevated risk
- Your fraud-risk factor list and the thresholds behind each one
- Your sampling policy — method, confidence level, tolerable error
Built for Claude
Install it into Claude Cowork and point it at your files.
- 1Download the skill bundle — everything it needs, including the shared building blocks, is inside.
- 2Add it to Claude Cowork as a plugin (audit-analytics).
- 3Point it at your own files and run it against one period you already know the answer to.
Free. We'll send the bundle and the setup notes.
Running on ChatGPT Enterprise or Copilot? We port these into your environment.
Want Journal Entry Testing & Fraud Risk Screening fitted to your methodology?
We adapt it to your control matrix, your workpaper template, and your thresholds — then hand it over so your team can run it themselves.