A command-line interface (CLI) based passive URLs discovery utility. It is designed to efficiently identify known URLs of given domains by tapping into a multitude of curated online passive sources.
Results extraction seems to be limited to 100 (i.e. the first page). Is necessary to check if the response contains more than 100 results and, if so, iterate to retrieve the other results.
Hi, As it stands, the URLScan check seems to me to be limited for two reasons :
Results extraction seems to be limited to 100 (i.e. the first page). Is necessary to check if the response contains more than 100 results and, if so, iterate to retrieve the other results.
Regards