The report discovers which classes of attacks are the most widespread regardless of the methodology used to identify them. Industry statistics such as those compiled by Mitre CVE project provide valuable insight into the types of vulnerabilities discovered in open source and commercial applications, this project tries to be the equivalent for custom web applications.
The statistics was compiled from web application security assessment projects which were made by the following companies in 2007 (in alphabetic order): Booz Allen Hamilton, BT, Cenzic with Hailstorm and ClickToSecure, dblogic.it, HP Application Security Center with WebInspect, Positive Technologies with MaxPatrol, Veracode with Veracode Security Review, WhiteHat Security with WhiteHat Sentinel.
The statistics includes 2 different data sets: automated testing results and security assessment results made using black and white box methodology.
Data analysis shows that more than 7% of analyzed sites can be compromised automatically. About 7.72% applications had a high severity vulnerability detected during automated scanning. Detailed manual and automated assessment using white and black box methods shows that probability to detect high severity vulnerability reaches 96.85%. So automated scanning represents data for an average Internet site and black and white box methods results refer to interactive corporate web applications.
The most prevalent vulnerabilities are Cross-Site Scripting, Information Leakage, SQL Injection and Predictable Resource Location. As a rule, Cross-Site Scripting and SQL Injection vulnerabilities appears due to system design errors, Information Leakage and Predictable Resource Location are often connected with improper system administration (for example, weak access control).