Snipwell

Jivaro Tool

Snipwell

Extract, clean, split, search, and export text from PDFs, scans, images, Office files, spreadsheets, web files, and pasted snippets. Everything runs in your browser.

Donate Get Proxies Get a VPN

Ready Add files to begin. Supported: PDF, images, DOCX, PPTX, XLSX, TXT, HTML, CSV, and Markdown.

Upload Command Center

Drop files, browse locally, paste images/text, or import a public URL from Advanced source.

0 files

Drop files here or choose files Private browser processing. OCR language data loads from a public CDN only when OCR is needed.

Browse files Add sample text

Advanced source: import from URL

Public file URL Works only when the remote site allows browser CORS access. Max remote file size: 50 MB.

Add from URL

Extraction Settings

Use Smart Extract for most files. Force OCR for scans and image-only PDFs.

Extraction mode Smart Extract OCR Everything Text Layer Only Mixed Page Repair

OCR language English English + Spanish English + French English + German English + Portuguese English + Italian English + Japanese Custom Tesseract codes

Advanced extraction controls

Smart OCR threshold

chars/page

OCR resolution scale

PDF pages

Max PDF pages/file

Page separator

Actions & Progress

Process selected files sequentially for better browser stability.

Extract Text Cancel Clear Files Reset Settings

Idle.

Load Read OCR Clean Ready

Clean Text

Fix common OCR and PDF extraction problems without changing the raw tab.

Clean OCR Light cleanup Preserve layout

Remove repeated headers/footers Remove standalone page numbers Join hyphenated words Repair broken line breaks Normalize spaces Normalize curly quotes

Apply Cleanup Restore Cleaned from Raw

Split Text

Create AI/message-ready chunks from the cleaned output.

Split by Characters Words Pages Headings Custom delimiter

Chunk size

Preserve whole sentences when practical Add “Message X of Y” labels

Build Chunks

Output Workspace

Review raw text, cleaned text, generated chunks, or extraction stats.

Raw Cleaned Chunks Stats

Search current output

Prev Next

No search term.

Export Center

Copy the active output or download cleaned text, chunks, and metadata.

Copy Current Text Download TXT Download MD Download JSON Download CSV

Privacy: files stay in your browser. URL imports depend on the source site allowing CORS. OCR can be slow for large scans.