The Hidden Dangers of Prototype Pollution: A Comprehensive Detection Framework
Authors:
Sumit Sharma (NIET)
Abstract

This research explores techniques for identifying prototype pollution vulnerabilities by sending requests that
trigger slight changes in server responses, proving the effectiveness of this detection method. Prototype
pollution is a significant cybersecurity issue and flaw that arises when an attacker manipulates JavaScript’s
prototype inheritance to alter an object’s prototype. Detecting this vulnerability on the server side in a black
box setting is challenging and can unintentionally cause denial-of-service incidents. Such alterations may allow
attackers to tamper with a "gadget" property, which could later be exploited in a vulnerable function.
This study confirms the possibility of safely detecting prototype pollution in a black-box environment by
observing subtle shifts in server behavior. Multiple approaches are used to automate the discovery of these
vulnerabilities. [29,11]

📄 Download Full Paper (PDF)
Published in: GCARED 2025 Proceedings
DOI: 10.63169/GCARED2025.p36
Paper ID: GCARED2025-0268