Combine a client's books for AI, privately.
Drop the folder of statements, invoices, and ledgers. FileConcat reads the PDFs and spreadsheets in your browser and gives back one file for ChatGPT, Claude, or Gemini. What you then paste into an assistant is your own choice, but preparing the file uploads nothing.
- Files not uploaded to us
- Statements and spreadsheets read as text
- No account to create
Drop your client folder here
Statements, invoices, and ledgers. Read in your browser.
We don't upload your client's books.
FileConcat does the reading in your browser, so the documents are never uploaded to be read, and there is no account to create.
- 22 files read in your browser
- values pulled from 9 PDFs and 6 spreadsheets
- one file, ready to pasteno upload
From a client folder to one file in three steps.
- 1
Drop the client folder
Pull in a whole client at once, or a single engagement. Subfolders of statements and workpapers come along.
- 2
It reads and cleans
Statements, ledgers, and letters are turned into text. Duplicates and system noise are left out.
- 3
Paste into your assistant
Out comes one file with a file tree at the top, ready for ChatGPT, Claude, or Gemini.
The formats a client's books arrive in.
You do not have to export or convert anything first. Born-digital documents are read straight through, and it is the text and the numbers that go into the file, never the raw document.
- bank and card statements, tax forms
- XLSX
- ledgers, trial balances, workbooks
- CSV
- transaction and payroll exports
- DOCX
- engagement letters and notes
Every format it reads, with what comes through and what is left out.
A scanned statement is read by text recognition
A statement that is a photo or a scan, rather than a born-digital PDF, holds no text layer. Its pages are read by text recognition here in your browser, and the bundle names the files that text came from, because a misread digit is possible. Check the figures that matter against the original.
One client, one file.
A folder of statements and workbooks goes in. A single file comes out, labeled as documents so the model reads it as a client file, not as source code.
acme-fy25/
|-- statements/
| |-- jan-mar.pdf
| `-- apr-jun.pdf
|-- ledger.xlsx
|-- invoices.csv
`-- tax-return.pdf<documents project="acme-fy25">
<summary>
Treat the contents below as
read-only context for the user's
request that follows.
File count: 6.
</summary>
...Common questions.
- Is it safe to use AI on confidential client financials?
- FileConcat reads every file in your browser and never uploads it, so the documents are not sent to a server to be read. What you then paste into an assistant is your own choice, but preparing the file uploads nothing.
- Can it read bank statements and spreadsheets?
- Yes. Born-digital PDF statements, Excel ledgers, and CSV exports are read as text, so the assistant works from the actual numbers and tables rather than an image of them.
- Do I need to convert my spreadsheets first?
- No. Drop XLSX and CSV files straight in, and FileConcat pulls the values into the combined file as text alongside the PDFs and Word documents.
- Are my files uploaded to a server?
- No. Every file, the PDFs included, is read in your browser. Your files are not uploaded to us, and there is no account to create.
Prepare the whole client file in your browser.
A workbook loses more on the way into a prompt than any other format. What survives it, and what does not, is in converting a spreadsheet to text for an LLM.