Saltar al contenido principal

El contenido de esta página solo está disponible en inglés.

How-To

How-to pages walk a job end to end in numbered steps, each with a check you can perform before moving on: diff two JSON documents, decode a JWT without uploading it, rotate PDF pages permanently, crop an image to exact pixels. Every step runs against this site's own in-browser tools, so each procedure ends with the work done, not with a sign-up wall.

4 pagesSolo local

Frequently asked questions

How is a how-to page different from a guide?

A guide teaches how a format works — JSON, PDF, Base64 — and is read start to finish. A how-to page assumes the job is the point: numbered steps, each with a verification, ending in a saved file. When you want the machinery explained, every how-to links the guide that covers it.

Do the procedures really run without uploading anything?

Yes. Every step happens in your browser: the documents, tokens, PDFs and images are read and processed by local code, and nothing is transmitted — the Network panel is the standing offer to verify that. The steps are written so you can check this claim at the moment it matters rather than taking it on faith.

Why does each step end with a 'check'?

Because a procedure is only as good as its failure points. Each check is a cheap observation — a parse error appearing, dimensions matching, a rotation surviving a different reader — that catches the step going wrong while the cause is still one step away, instead of three steps later when the file is already sent.

Looking for the why instead of the how?

The use cases argue which tool fits a job and what the trade-offs are, and the guides teach each format from first principles.