txtcmp is tool for finding similar text files. It is meant for the case where
you have many files and a few may be similar. It works by computing the longest
common subsequence (LCS) of all provided files, which is what is at work in
utilities like diff. It is not as useful in the case where you have just two
files. You should use diff for that.
Compare three files
txtcmp file1.txt file2.txt file3.txt
Find the most similar files in a directory
find path/to/dir -type f | xargs txtcmp | sort -n | tail