Base32 decode
Base32 decode. Runs on your device, nothing is uploaded.
Base32 decode reads standard RFC 4648 text, A through Z and 2 through 7, drops whitespace and trailing = padding, then rebuilds the bytes five bits at a time. An out-of-alphabet character halts the run, named by its position in the cleaned string. UTF-8 comes back as text; anything else downloads as decoded.bin so nothing looks corrupted.
Questions
What alphabet does it expect?
The RFC 4648 base32 alphabet, A through Z and 2 through 7, case-insensitive.
Does padding matter?
No, trailing = padding is stripped before decoding, so input with or without it works the same.
What happens with an invalid character?
The tool refuses, naming the character and its offset in the cleaned string.
What does the output look like?
Decoded UTF-8 text prints directly; anything else downloads as decoded.bin.
Can I use this from the command line?
Yes, the same code runs on the page, the CLI and MCP.