The text is turned into UTF-8 bytes and those bytes are encoded, which is the step the short snippets skip. The summary tells you how many bytes went in, and that is not the same as the number of characters you typed: anything outside plain ASCII takes two to four bytes on its own, so a line of Turkish or Japanese counts higher than it looks.
This page loads no libraries at all. The work is done by the browser itself. Your file is read on your own machine and never sent anywhere. Open your browser’s network tab and watch while you use it.
The browser’s btoa function accepts Latin-1 and nothing above it. A character like 東 or an emoji makes it throw an invalid character error outright, and an accented letter goes in as a single Latin-1 byte that reads as something else at the far end: é encodes as 6Q== there instead of w6k= here. Going through UTF-8 bytes first avoids both.
No. This is the standard alphabet, which uses plus and slash, and it is padded with equals signs. A query string or a JWT wants the URL-safe variant, with minus and underscore and no padding, and that is not what comes out here.
Check the end of your text first. Nothing is trimmed here, so a trailing newline or space is encoded along with the rest and changes the tail: hello gives aGVsbG8= and hello with a newline after it gives aGVsbG8K. The output is also one unbroken line, without the breaks mail formats put in every 76 characters.
No, and treating it as such is the mistake worth avoiding. It writes bytes using a set of 64 characters that survive being copied around, and anyone holding the string can turn it back. Use it to carry text through something, never to hide it.
Everything here is free and works the same way, in your browser.
A new DeskKit tool lands every week or two. Leave an address to hear about it. Nothing else is ever sent, and one click removes you.