Fileless cURL to Chromium

less than 1 minute read

Sometimes (expecially during CTFs) I need to display the result of a REALLY specific HTTP request that I made with cURL into Chromium.

The naive and boring way of doing this would be something like this:

curl -s > /tmp/page.html
chromium /tmp/page.html
rm /tmp/page.html

For some reason, I got stubborn on not creating that temporary file, which led to the mess you can see below.


curl -s URL | base64 -w 0 | xargs -i chromium "data:text/html;base64,{}"


The biggest issue is that chromium can’t open files from stdin, but only from URLs passed as argument. The trick here consists in passing a url with the “data:” schema, appending the actual content it should display.

  • curl -s URL: makes the actual request. The -s switch will not display the annoying “elapsed time” stuff when output is redirected.
  • base64 -w 0: encodes stdin in Base64 and passes the result to stdout. -w 0 disables line wrapping after 76 cols, which stops base64 from inserting newline characters.
  • xargs -i "something {}": it takes what is passed in stdin and places it where {} is.