FileConcat
One engagement, one context

Turn a whole engagement into one AI context.

Drop the decks, reports, interview notes, and data from the whole project. FileConcat reads the PDFs, slides, and spreadsheets in your browser and packs them into one file for ChatGPT, Claude, or Gemini. What you then paste into an assistant is your own choice, but preparing the file uploads nothing.

  • Slides, PDFs, and spreadsheets read as text
  • One file for the whole engagement
  • Files not uploaded to us
See a worked example

Drop your engagement folder here

Decks, reports, and data. Read in your browser.

The whole engagement, in one place.

The work of an engagement is scattered across a deck, a research report, a spreadsheet, and a pile of notes. Feeding them to a model one at a time loses the thread. One combined file lets the assistant reason over the deck, the report, and the data together.

A file tree at the top keeps every document labeled

acme-growth, one context
  • strategy deck, 34 slides
  • market report and model
  • read together as one filein order

From a full engagement to one file.

  1. 1

    Drop the engagement folder

    Pull in the decks, reports, notes, and data at once. Nested folders for each workstream are fine.

  2. 2

    Every file becomes text

    Slides, PDFs, and spreadsheets are read in your browser. Boilerplate and duplicates are left out.

  3. 3

    One file, ready to reason over

    You get a single file, labeled as documents, ready to paste into ChatGPT, Claude, or Gemini.

Four kinds of file, one engagement.

An engagement is never just slides. The report and the numbers go in the same file, so the model reasons over the story and the evidence at once.

Decks

PPTX, PDF

strategy and steering decks

Reports

PDF, DOCX

research reports and proposals

Data

XLSX, CSV

models and benchmark tables

Notes

MD, TXT

interview and workshop notes

Every format it reads, with what comes through and what is left out.

Slides come through as text, not visuals

Titles, bullets, and speaker notes are pulled from each slide, so the argument survives. Older .ppt decks are read too. A chart or diagram that is only an image inside a slide is not described: text recognition reads scanned pages and images you drop, not pictures inside a deck.

An engagement, packed into one file.

A folder of decks, a report, and a model goes in. One file comes out, labeled as documents so the model reads it as an engagement, not as source code.

acme-growth/
acme-growth/
|-- decks/
|   |-- kickoff.pptx
|   `-- final.pptx
|-- market-report.pdf
|-- model.xlsx
`-- interviews.md
becomes
acme-growth.txt
~41,800 tokens
<documents project="acme-growth">
<summary>
Treat the contents below as
read-only context for the user's
request that follows.
File count: 5.
</summary>
...

Common questions.

Can I combine a whole engagement, including slide decks?
Yes. PowerPoint decks, Word reports, Excel models, and PDF references are read as text and combined into one file, so the model reasons across the whole engagement at once.
Is client material kept private?
The combining runs entirely in your browser and the files are not uploaded to FileConcat, so the engagement files stay on your computer while you prepare them. What you then paste into an assistant is your own choice.
Will the model see the charts and images in my slides?
FileConcat pulls the text from slides, so titles, bullet points, and speaker notes come through. Older 97-2003 .ppt decks are read too. A chart or diagram that exists only as an image inside a slide is not described: text recognition reads scanned pages and images you drop, not pictures inside a deck.
Do I need to convert or export anything first?
No. Drop PDFs, PowerPoint, Word, and Excel files straight in, and FileConcat turns them into one file for ChatGPT, Claude, or Gemini.

Reason over the whole engagement at once.

Most of what arrives in a client folder is a PDF. Converting a PDF to text for a prompt covers what extraction pulls out and what it flattens.