For CFOs, controllers, and fund accountants: 10 prompts plus an offline extractor that match scanned bank, custodian, and brokerage statements to your ledger in seconds and return only the breaks, on your own machine. An alternative to BlackLine.
Free — runs in your own ClaudeMedium setup · 5 steps10 ready-to-run prompts
Three minutes, four steps, nothing to install by hand
Claude sets it up for you. You just paste.
Never used Claude? It is free and takes 30 seconds to open. Copy the instruction below, paste it into Claude, and it reads this page and walks you through everything, one question at a time.
1
Tell Claude how to talk to you
One tap. It changes how much Claude explains, and how slowly it goes. You can change it any time.
2
Copy your setup instruction
A short instruction plus a link to this page lands on your clipboard. First copy asks for your email once. That unlocks every button across the whole library.
3
Open Claude in a new tab
Free account, no card, 30 seconds. This tab stays open so you can come back.
Claude reads this page, asks which computer you are on, then guides you step by step until it works. If anything errors, tell Claude what you see, and it fixes it with you.
▸Prefer the full prompt instead of the link? (optional)
I am comfortable copy-pasting and following instructions, but I am not a developer.
- Plain English. Define jargon the first time it appears.
- One step at a time, then wait for me to confirm before the next one.
- Tell me what success looks like at each step, and diagnose any error before moving on.
Follow the instructions below with those rules applied.
I want to set up the Statement Matching Pack and reconcile my first real statement against my ledger before the close. Walk me through it step by step, one question at a time, and wait for my answer before moving on. I am a CFO, controller, or fund accountant, not a developer. Define every term once. Keep me to one action per message.
## What this produces
A statement matched to my ledger in seconds, with only the breaks coming back for me to review: duplicates, timing differences, missing wires, amount mismatches, each with a reason. The extraction runs offline on my own machine, so the statements never leave the building. A footing proof at the end shows the reconciliation actually ties before anything books.
## Step one, ask me this first, nothing else
"Is the statement you want to match a clean digital export (CSV or native PDF), or a scanned or photographed PDF? Either works. I just need to know whether we turn on the offline reader for scans."
Wait for my answer. Then continue, one step per message:
## The path you walk me down
- Classify my setup as CLI plus chat (we install the offline reader once, then everything else is in Claude). Tell me the one command to install it, where to paste it, and what success looks like.
- Have me extract one statement with a single command, then check the extracted closing total against the printed total on the statement before we go further. Stop me if it does not match and help me fix the extraction.
- Walk me through creating a Claude Project named "Recon Desk" and pasting the house rules (match on amount, date, reference; never invent a transaction; never mark a break resolved on your own; mask account numbers to last four; list everything that does not match and why).
- Have me load the two inputs: the extracted statement rows and my ledger export.
- Run the first match with me on a small set, then read the break list together so I trust it before I run the full statement.
- Set my tolerances (how many days counts as a timing difference, how many cents counts as a near match) from what my own data shows, not a guess.
- Finish with the footing proof: opening plus matched activity plus or minus breaks equals closing. If it does not foot, name the exact gap.
## The first session drill
By the end of this session I should have: one real statement extracted offline, matched against my ledger, a break report with every account number masked to last four, and a footing proof that ties. Then show me the one line command and the prompt order so next month's close is a checklist, not a rebuild.
Before you tell me the reconciliation is done, re derive the matched total and the break total, confirm matched plus unmatched equals the total on both sides, and confirm it foots. Do not say it is ready until that check is clean. I review every break and sign the reconciliation myself.
Step 2 · run it on your data
Step 1 set it up. These 10 prompts do the work.
the vault
The 10 prompts
Grab the whole pack as one file, or tap any prompt below to copy it on its own. Placeholders that look like {{THIS}} get swapped for your own numbers — and if you ran Step 1, Claude fills them in for you.
One .md file · all 10 prompts, numbered, in order · nothing left out.
Here are the extracted rows from a bank or custodian statement, parsed offline. Confirm every row has a date, a description, and an amount. Flag any row where a number looks mis read: an amount with no decimal, a date out of range, a payee that looks truncated. Then tell me the extracted closing total so I can check it against the printed statement total before we match anything.
Match each transaction in the extracted statement to a line in the ledger I loaded. Match on amount first, then date within {{TOLERANCE_DAYS}} days, then reference. Return a table of matched lines (statement line, ledger line, what tied them), then a separate list of every statement line with no ledger match and every ledger line with no statement match. Do not force a match you are not sure of. End with the count: matched, unmatched statement lines, unmatched ledger lines, and confirm they sum to the totals on both sides.
Scan the statement for likely duplicate transactions: same or near same amount, same payee, within {{WINDOW}} days. List each suspected pair with the reason it looks like a duplicate and the reference on each. Do not assume a duplicate, flag each pair for my review.
For every unmatched line, classify it: timing difference (a match exists in the other set within {{TOLERANCE_DAYS}} days, just outside the period), true break (no match anywhere), or unclear. For each timing difference, show the candidate matching line so I can confirm it. Leave only genuine breaks on the true break list.
List every transaction that appears on exactly one side, the statement or the ledger, not both. Give amount, date, and payee for each, sorted by amount largest first, and state which side it is missing from. These are the one sided items I work first.
Produce a break report from the matching. Section one: summary (total lines, matched, breaks, total dollar value of breaks). Section two: a table of every break with amount, date, payee, type (duplicate, timing, missing, amount mismatch), and a one line suggested next step. Mask every account and routing number to the last four digits. Keep matched plus broken equal to the totals from prompt 02.
Find pairs that match on payee and date but differ in amount by less than {{CENTS}} cents or less than {{PERCENT}} percent. List each with the exact difference and a likely cause (bank fee, FX, rounding). Keep these near matches separate from true mismatches so I can set the tolerance myself.
Some statement lines are in {{CURRENCY}}. Match them to the ledger using the rate I provide ({{RATE}}, or a rate column in the file). On every cross currency match, show the original amount, the converted amount, and the rate used. Flag any line where the implied rate is more than {{PERCENT}} percent off the rate I gave.
Prove the reconciliation foots. Opening balance plus matched activity plus or minus identified breaks should equal the closing balance on the statement. Show the arithmetic line by line. If it does not foot, tell me the exact gap. Do not round it away.
Save the matching rules we settled on (tolerance days, cents tolerance, duplicate window, currency handling) as a short standard I can paste at the start of next month's run. Then give me the one line extraction command and the order of prompts to run, so next close is a checklist, not a rebuild.
For CFOs, controllers, and fund accountants: 10 prompts plus an offline extractor that match scanned bank, custodian, and brokerage statements to your ledger in seconds and return only the breaks, on your own machine
Path A · free
You just did it
The setup rail and every prompt above are free and stay free. The cost is your time, and the risk of wiring it wrong on live data.
• Run last quarter's numbers first. Live data is not a test bed.
• Nothing here uploads to us. It runs in your own Claude account, on your own machine.
• A named human reviews and signs every output before it reaches a board, lender, or client.
• Wiring the open-source piece to real systems? Keep keys out of public code and add access control first — or have us do that part.
the fine print
Credit the original author
Prompt set authored by consultance.ai. MinerU is Apache 2.0 and runs offline on your machine. No affiliation with BlackLine or Trintech implied. The statements never leave your machine; a named human reviews every break and signs the reconciliation. Not accounting, audit, or financial advice.
Want this running in your business, not just your laptop? We build it and hand you the keys.
Statement Matching and Reconciliation is a finance and data build in the consultance.ai AI Build Library. For CFOs, controllers, and fund accountants: 10 prompts plus an offline extractor that match scanned bank, custodian, and brokerage statements to your ledger in seconds and return only the breaks, on your own machine. An alternative to BlackLine. It fits CFOs, controllers, fund accountants, and PE or family office finance teams whose month or quarter end is days of matching scanned bank, custodian, and brokerage statements to the ledger line by line. Setup difficulty is Medium, with 5 plain-English steps.
What does Statement Matching and Reconciliation do?
For CFOs, controllers, and fund accountants: 10 prompts plus an offline extractor that match scanned bank, custodian, and brokerage statements to your ledger in seconds and return only the breaks, on your own machine. An alternative to BlackLine.
Who is Statement Matching and Reconciliation for?
It fits CFOs, controllers, fund accountants, and PE or family office finance teams whose month or quarter end is days of matching scanned bank, custodian, and brokerage statements to the ledger line by line.
How hard is Statement Matching and Reconciliation to set up?
Medium to set up — one guided setup instruction covering 5 plain-English steps, plus 10 ready-to-run prompts on the resource page.
How would consultance.ai build this out?
We would deliver a deployed recon desk: the offline extractor installed and tuned to your statement formats, the prompts loaded into your Claude tenant, your matching tolerances set from your own history, a break report template mapped to your reviewers, and a compliance overlay for your sign off. Done with you, then handed over so you own it and the data never leaves your machine.
What are the licensing terms?
Prompt set authored by consultance.ai. MinerU is Apache 2.0 and runs offline on your machine. No affiliation with BlackLine or Trintech implied. The statements never leave your machine; a named human reviews every break and signs the reconciliation. Not accounting, audit, or financial advice.
Want this built into your workflow?
Statement Matching and Reconciliation is the starting point. On a free AI audit we map where it fits your stack and what consultance.ai would build around it.