cat file > tr -d '\r'
cat file > rm

cat file | tr -d '\r'         # tr reads stdin
cat file | xargs -d '\n' rm   # rm reads arguments

You are using file redirection, but the filename is an unquoted command name. Instead of running the command and feeding data to it, this just writes to a file with the same name.

To run the command and feed data to it, determine how it gets its data:

• If the command reads from STDIN, simply use a pipe as in the first example.
• If the command reads multiple arguments, use a pipe to xargs as in the second example

Note that xargs has many pitfalls when it comes to spaces and quotes. cat file | xargs rm will appear to work during testing, but fails for filenames like My File.txt or Can't_Fight_This_Feeling.mp3. The example uses the GNU extension -d '\n' to more safely handle these names.

If you actually did want to write a file named after a command, simply quote the filename to let ShellCheck know you meant it literally and not as a command name. This does not change anything about how the script works:

# Write to a file literally named 'rm', does not try to delete anything
echo "A potentially dangerous command" > "rm" 

• Help by adding links to BashFAQ, StackOverflow, man pages, POSIX, etc!