Base64 is not encryption: encode Unicode text correctly
Base64 turns bytes into a text representation. It is handy when a system expects text, but the transformation is reversible without a password. If someone can read the encoded text, they can usually recover the original bytes.
Try a reversible example
Open the Base64 tool and encode Hello. The result is SGVsbG8=. Paste that result into the input and choose Decode to get the original word back. Copying the encoded value into a public issue tracker therefore does not hide the content.
Why Unicode needs a byte encoding first
English letters, accented characters and emoji do not all have the same UTF-8 byte length. ToolDock converts text to UTF-8 bytes before Base64 encoding, and decodes the bytes as UTF-8 on the return trip. Try Hello 👋 or café and check that decoding preserves the characters. The MDN Base64 overview explains the distinction between bytes and JavaScript text.
What this tool accepts
This is a text encoder and decoder. It is not a file decoder, image extractor or encrypted-secret manager. If Base64 decodes to bytes that are not valid UTF-8, it reports an error instead of pretending those bytes are readable text. Do not remove random characters from the input just to suppress that error; find out what kind of data the producer actually encoded.
Standard Base64 and Base64url are different
This tool uses standard Base64. Some protocols use a URL-safe variant with a different alphabet and padding conventions. A token segment from such a protocol may not work unchanged here. Encoding a URL parameter is also a different task: use the URL Encoder when you need percent-encoding for a query value.
Keep secrets out of shared examples
Use a harmless sample when reporting a problem. Do not share an encoded password, session cookie or authorization header under the assumption that Base64 conceals it. Local processing means this tool does not upload your input; it does not make an exposed credential safe.