How does one delete the first known character in a string with sed?

For example, say I want to delete the first character (which is @) in the string "@ABCDEFG1234"

I want sed to verify whether "@" exists as the first character. If so, it should delete the character.

3 Answers 3


sed 's/^@\(.*\)/\1/'

^ means beginning of the string

@ your known char

(.*) the rest, captured

then captured block will be substituted to output Sorry, can't test it at the moment, but should be something like that

  • I want to delete + which does not need to be escaped although it stands for 1 or more in regex.
    – Timo
    Nov 4, 2020 at 7:56

There's no need to capture and replace.

sed 's/^@//'

This replaces the character @ when it's first ^ in the string, with nothing. Thus, deleting it.


You can do this instead.

sed 's/^.//'

^ - Starting character 
. - No of charactes(.. means two characters)

Example :

echo test123 | sed 's/^.//'
echo test123 | sed 's/^..//'
  • 2
    The question says “I want sed to verify whether "@" exists as the first character. If so, it should delete the character.” Your answer deletes the first character of each line no matter what is it. For example, if a line contains test123, it looks like the OP wants it left alone. Mar 12, 2018 at 3:57
  • I like the quick solution of Prakaash, but @Scott you are right when reading the OP carefully. In most of the cases, you do want a distinct character and not any.
    – Timo
    Nov 4, 2020 at 7:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .