Batch processing
Run one operation over a whole folder
Point it at a folder of 4,000 files and come back in ten minutes. Every upload site on the internet is one file at a time, because asking you to upload four thousand scanned invoices is unthinkable for them. Here there is nothing to upload: pick an operation, preview the first real result, then let it run. No file limit, no size limit — your machine is the limit.
1. Choose inputs
This browser provides folders read-only; completed results come back as a ZIP.
Drop any file or paste text to launch tools instantly
Auto-detects PDF, JSON, CSV, Images, Timestamps, SQL, Colors, and URLs. 100% processed in your browser memory.
No files selected.
Chain steps (optional)
Scan, deskew, OCR, redact, compress, sign — one drop, every step, in order. A hosted tool makes you download the result and pick it up again between each step, because each step is a separate job on someone else’s machine that someone has to bill for. Here the file simply stays in memory from one step to the next.
Build the chain
Each step runs on whatever the one before it produced. Until you add a step, the single-operation run below stays unchanged.
Add the first step
648 operations can start a pipeline.
Validate ascending chapter timestamps and normalize a paste-ready chapter list.
2. Pick the operation
Word counter
Count the words in a passage using Unicode letter and digit runs, so accented words count once and a contraction such as don’t stays a single word.
3. Preview, then run
Tokens: {name} {ext} {index} {operation} {date}