Audit sampling
Create a reproducible sample from a messy journal export.
Turn variable accounting journals into audit samples a reviewer can replay.
Give it a readable journal and sampling method. Get back a normalized population, selected sample, seed, filters and reading checks.
What it does
From mixed journals to a reviewable sample
Use it when a journal is readable but not standardized: good Excel, CSV, print-style Excel exported from PDF, or text PDF. Codex asks only for the few field choices that cannot be inferred safely.
input/
journal.xlsx
print_export_from_pdf.xlsx
general_journal.pdf
out/
inspection.json
suggested_recipe.json
normalized_journal.csv
normalization_diagnostics.json
sample/
journal_sample.csv
journal_sample.xlsx
sampling_audit.json
Inputs
What it reads
Works with local files already available to the firm. OCR-only scanned PDFs are flagged for manual handling rather than treated as readable input.
Outputs
What it produces
The result is a reviewable work folder: recipe, normalized population, sample files, reading checks and review trail.
Sample selection you can replay
The inspection step proposes the accounting columns and asks only for essential unresolved choices: header row, account column, debit/credit convention or sample assumptions.
The resulting folder keeps extraction, normalization, filters, sample seed and exports visible, so a reviewer can repeat or challenge the selection.
Reading gaps stay visible instead of being hidden in the final sample.
Use
Create the sample from the work folder
Open the work folder, confirm the essential field choices and generate the sample package.
Prompts
Ready prompts
Start from one of these prompts and replace the folder path, language, method and sample size. Codex will ask before resolving essential ambiguities.
Installation
Install from Plugin Pack
Requires Codex desktop. One ZIP installs all Mparanza plugins.
First use
In a Codex chat, provide the journal folder, language, sampling method and target size.
Use Journal Sampling on /path/input.
Language: en. Document language: auto.
Inspect the files, ask only for essential field ambiguities, then generate sample, reading checks and review trail.
Method: random. Size: 25.