Measurements, Attacks, and Defenses for the Web (MADWeb) Workshop 2022

Billions of people use Chrome to go about their daily lives: for school, for work, for joy, for shopping, for curiosity, and for secrets. Members of the Chrome engineering team are deeply conscious of (and constrained by) the responsibility that comes with broad product adoption. Adriana will talk about the lessons learned through a decade on the Chrome team. She’ll share surprises that she encountered when she transitioned from academia to industry, best practices that she learned as a Chrome security team member, and challenges that she has faced as a product engineering leader. What do you do when security goals conflict with other user needs? How do you balance execution speed with security? What’s hard to do even when you have Google's resources behind you? Where is academia ahead of industry, and what papers are sitting on engineers' desks?

Speaker's Biography

Adriana Porter Felt is a Director of Engineering at Google, where she leads Chrome’s Data Science, content ecosystem, and iOS teams. Previously, Adriana founded and led Chrome's usable security team. She is best known externally for her work on moving the web to HTTPS, earning her recognition as one of MIT Technology Review's Innovators Under 35. Adriana holds a PhD from UC Berkeley, and most of her academic publications are on usable security for browsers and mobile operating systems.

On GitHub, open-source developers use the fork feature to create server-side clones and implement code changes separately before creating pull requests. However, such fork repositories can be abused to store and distribute malware, particularly malware that stealthily mines cryptocurrencies.

In this paper, we present an analysis of this emerging attack vector and a system for catching malware in GitHub fork repositories with minimal human effort called Fork Integrity Analysis, implemented through a detection infrastructure called Fork Sentry. By automatically detecting and reverse engineering interesting artifacts extracted from a given repository’s forks, we can generate alerts for suspicious artifacts, and provide a means for takedown by GitHub Trust & Safety. We demonstrate the efficacy of our techniques by scanning 68,879 forks of 35 popular cryptocurrency repositories, leading to the discovery of 26 forked repositories that were hosting malware, and report them to GitHub with seven successful takedowns so far. Our detection infrastructure allows not only for the triaging and alerting of suspicious forks, but also provides continuous monitoring for later potential malicious forks. The code and collected data from Fork Sentry will be released as an open-source project.

Uniform Resource Identifier (URI) schemes instruct browsers to conduct specific actions depending on the requested scheme. Previous research has addressed numerous issues with web URI schemes (e.g., http: and https:) both for desktop and mobile browsers. Less attention has been paid to local schemes (e.g., data: and file:), specifically for mobile browsers. In this work, we examined the implementation of such schemes in Android OS browsers, analysing the top-15 mobile browsers. As a result, we discovered three vulnerability types that affect several major browsers (including Google Chrome, Opera and Samsung Internet). First, we demonstrate an URI sanitisation issue that leads to a cross-site scripting attack via the JavaScript scheme. The problem affects Chromium browsers including Chrome, Opera, Edge, and Vivaldi. Second, we found a display issue in Samsung Internet that allows abusing data URIs to impersonate origins and protocols, posing a threat in the context of phishing attacks. Finally, we discover a privilege escalation issue in Samsung’s Android OS, leading to full read and write access to the internal storage without user consent and bypassing the Android storage permission. While this issue was originally discovered in the file scheme of the Samsung browser, utilising a combination of static and dynamic analysis, we traced the problem back to an authorization issue in Knox Sensitive Data Protection SDK. We then show that any app can abuse this SDK to obtain full access to the internal storage without appropriate permission on Samsung devices running Android 10. We responsibly disclosed the vulnerabilities presented in this paper to the affected vendors, leading to four CVEs and security patches in Chrome, Opera and Samsung Internet browser.

Keywords—Android, mobile browsers, XSS, privilege escalation, URI schemes

The log4shell vulnerability has been called one of the most significant cybersecurity vulnerabilities in recent history. For weeks after initial disclosure, companies around the globe scrambled to respond by patching their systems or by applying mitigating security measures to protect systems that could not be readily patched. There are many possible ways to detect if and where an organization is vulnerable to log4shell, each with advantages and disadvantages. Penetration testing in particular is one possible solution, though its results can be misleading if not interpreted in the proper context. Mitigation measures have varying degrees of success: Web Application Firewalls (WAFs) could be bypassed, whereas our analysis revealed that outbound network restrictions would have provided an effective protection given the rapidly evolving patch cycle. Ultimately, log4shell should change the way we look at web attack surfaces; doing so will ensure we can be better prepared for the next critical zero-day Remote Code Execution (RCE) vulnerability.

Index Terms—Log4j; Web Attack Surface; Penetration Testing

Online advertisements are an unavoidable fact of the modern web: they are embedded in and financially support the majority of content websites. Significant prior work in the computer security and privacy community has previously studied the ecosystem of online advertising, particularly in terms of its privacy implications. What has not been substantively considered in the security community, however, is the visible, user-facing content of these advertisements. Our recent work reveals significant prevalence of a range of problematic content in these ads, including clickbait, misinformation, scams, and manipulative design patterns. In this talk, I will describe our work characterizing and measuring problematic content in the online ad ecosystem, including an investigation of ad content on misinformation sites and a study of political-themed ads on news and media websites around the time of the 2020 U.S. elections.

Speaker's Biography

Franziska (Franzi) Roesner is an Associate Professor in the Paul G. Allen School of Computer Science & Engineering at the University of Washington, where she co-directs the Security and Privacy Research Lab. Her research focuses broadly on computer security and privacy for end users of existing and emerging technologies. Her work has studied topics including online tracking and advertising, security and privacy for sensitive user groups, security and privacy in emerging augmented reality (AR) and IoT platforms, and online mis/disinformation. She is the recipient of a Consumer Reports Digital Lab Fellowship, an MIT Technology Review "Innovators Under 35" Award, an Emerging Leader Alumni Award from the University of Texas at Austin, a Google Security and Privacy Research Award, and an NSF CAREER Award. She serves on the USENIX Security and USENIX Enigma Steering Committees.

Services on the public Internet are frequently scanned, then subject to brute-force password attempts and Denial-of-Service (DoS) attacks. We would like to run such services stealthily, where they are available to friends but hidden from adversaries. In this work, we propose a discovery-resistant moving target defense named “Chhoyhopper” that utilizes the vast IPv6 address space to conceal publicly available services. The client meets the server at an IPv6 address that changes in a pattern based on a shared, pre-distributed secret and the time of day. By hopping over a /64 prefix, services cannot be found by active scanners, and passively observed information is useless after two minutes. We demonstrate our system with the two important applications—SSH and HTTPS, and make our system publicly available.

One of the main roles of the Domain Name System (DNS) is to map domain names to IP addresses. Despite the importance of this function, DNS traffic often passes without being analyzed, thus making the DNS a center of attacks that keep evolving and growing. Software-based mitigation approaches and dedicated state-of-the-art firewalls can become a bottleneck and are subject to saturation attacks, especially in high-speed networks. The emerging P4-programmable data plane can implement a variety of network security mitigation approaches at high-speed rates without disrupting legitimate traffic.

This paper describes a system that relies on programmable switches and their stateful processing capabilities to parse and analyze DNS traffic solely in the data plane, and subsequently apply security policies on domains according to the network administrator. In particular, Deep Packet Inspection (DPI) is leveraged to extract the domain name consisting of any number of labels and hence, apply filtering rules ( e.g., b locking malicious domains). Evaluation results show that the proposed approach can parse more domain labels than any state-of-the-art P4-based approach. Additionally, a significant performance gain is attained when comparing it to a traditional software firewall -pfsense-, in terms of throughput, delay, and packet loss. The resources occupied by the implemented P4 program are minimal, which allows for more security functionalities to be added.

Index Terms—P4-programmable switches, stateful processing, high-speed networks, DNS filtering, DPI.

In this paper we perform the first empirical analysis of the use of web storage in the wild. By using dynamic taint tracking at the level of JavaScript and by performing an automated classification of the detected information flows, we shed light on the key characteristics of web storage uses in the Tranco Top 5k. Our analysis shows that web storage is routinely accessed by third parties, including known web trackers, who are particularly eager to have both read and write access to persistent web storage information. This motivates the need for further research on the security and privacy implications of web storage content.

While security researchers are adept at discovering vulnerabilities and measuring their impact, disclosing vulnerabilities to affected stakeholders has traditionally been difficult. Beyond public notices such as CVEs, there have traditionally been few appropriate channels through which to directly communicate the nature and scope of a vulnerability to those directly impacted by it. Security.txt is a relatively new proposed standard that hopes to change this by defining a canonical file format and URI through which organizations can provide contact information for vulnerability disclosure. However, despite its favourable characteristics, limited studies have systematically analyzed how effective Security.txt might be for a widespread vulnerability notification campaign. In this paper, we present a large-scale study of Security.txt’s adoption over the top 1M popular domains according to the Tranco list. We measure specific features of Security.txt files such as contact information, preferred language, and RFC version compliance. We then analyze these results to better understand how suitable the current Security.txt standard is for facilitating a large-scale vulnerability notification campaign, and make recommendations for improving future version of the standard.