Private Document Search
Search across your own PDFs, Word files and spreadsheets without uploading them.
Drop the files you want to search
PDF, Word, Excel, CSV, text, Markdown and HTML. Read in this tab only — nothing is uploaded, and closing the tab forgets them.
How this works, and what it cannot do
Each file is read in your browser and split into passages of a few sentences. Your search is ranked with BM25 — the method behind most search engines — so rarer, more specific words count for more. English words match their other forms, and Japanese is searched by overlapping character pairs. Use "quotes" for an exact phrase and a minus sign to exclude a word.
The highlighted “best match” is quoted from your file, not written by an AI, so it cannot make anything up — but it also cannot combine facts from two places into one answer. Scanned PDFs need OCR first. Nothing is uploaded or stored: close the tab and the index is gone.
Private by design. This tool runs entirely in your browser. Your file is processed on your own device and is not uploaded by DO101.
About the Private Document Search
Drop in the files you need to search — PDFs, Word documents, Excel sheets, notes, exports — and search across all of them at once. Results are ranked with the same method most search engines use, point to the exact passage that matched with the words highlighted, and the best-matching sentences are quoted at the top with their source. It understands different forms of a word and searches Japanese too. Everything is read and indexed inside this browser tab: nothing is uploaded, nothing is stored, and closing the tab forgets it all. No AI service, so it quotes your documents rather than paraphrasing them.
How to use it
- 1Drop in PDFs, Word documents, spreadsheets or text files.
- 2Type a question or keywords. Use "quotes" for a phrase and -word to exclude.
- 3Read the quoted best match, then the ranked passages below it.
- 4Filter to one file, or copy an answer with its source.
What you get
- Searches PDF, Word, Excel, CSV, text, Markdown and HTML together
- Ranked by BM25, the method behind most search engines
- Points to the exact passage, with matches highlighted
- Quotes the best-matching sentences with their source
- Matches other forms of a word (invoice, invoicing)
- Searches Japanese text
- Exact phrases and excluded words
- Nothing uploaded; closing the tab forgets everything
Frequently asked questions
Are my files uploaded?
No. They are opened and indexed inside this browser tab. There is no server receiving them, and closing the tab discards the index.
Can I ask it questions like a chatbot?
You can type a question, and it will find and quote the passages that best match it. It does not write a new answer in its own words — that needs an AI service. The upside is it cannot invent anything: every answer is a quote from your file.
Why is a PDF not searchable?
Scanned PDFs are images with no text in them. Run them through the PDF OCR tool first, then search the result.
How many files can I search?
As many as your browser can hold in memory — typically hundreds of ordinary documents. Very large files take a few seconds to read.
Does it search Japanese?
Yes. Japanese has no spaces between words, so it is indexed by overlapping character pairs, the way search engines handle it without a dictionary.
Keep going
People who use this tool usually reach for these next.
Pull the plain text out of a PDF.
Read the text out of a scanned PDF.
Answer repeat customer questions from your own FAQ, in English or Japanese.
Back from time off? See the messages that matter to you first, and why.