Are there better alternatives to null-terminated strings?

a variable-length integer encoding that somewhat resembles what they do in UTF-8. It means for strings < 128 chrs, the length is a single byte. Longer than that and more bytes get used as necessary.

What you used might be similar to unsigned LEB128, which is used in DWARF, Webassembly, Android's DEX format, and protobuf. Essentially encodes 7 bits of the number in each byte, with the high bit being 1 in any byte except the last one representing the number.

Though unlike UTF-8 the number's length isn't encoded in the first byte but instead implied by the final byte. Arguably making the number's encoding similar to a terminated string.