// Learning Phase and Response Filtering
Identify and fix each vulnerability on your website and web application thanks to a patch without modifying code, by identifying different types of contents.
Try it free
One of the hardest things to do during the learning phase is to identify all query string and body parameters having a well-known format. This refers to: “numeric only”, alphabetical, alphanumeric, etc.. among all content types requested such as form urlencoded, JSON, XML, multipart, and many others.
How does Mithril Learning Phase and Response Filtering work?
Mithril has a learning module that automatically analyzes the customer’s HTTP traffic, and creates custom Web Application Firewall rules to limit and sanitize the user’s input. Sometimes this is not enough to patch misconfigurations or vulnerabilities in a web application.
Therefore, our team is able to apply virtual patches by reading a Penetration Test Report or after doing a Penetration Test activity. This makes us able to customize WAF Rules and Modules to sanitize input even replacing on-fly any response body to redact information leakage and debug.
Just an example
Your web application has a parameter name “newsid”, and its content is just an int that refers to a news id number on your database. Our learning module will automatically produce a rule having “ARGS:newsid” as a variable checking its value with a regular expression like “^[0-9]+$”. If the regex doesn’t match, Mithril will block the request filtering out all injection attempts like SQL Injection.