Given the dynamic nature of the Web, security measurements on it suffer from reproducibility issues. In this paper we take a systematic look into the potential of using web archives for web security measurements. We first evaluate an extensive set of web archives as potential sources of archival data, showing the superiority of the Internet Archive with respect to its competitors. We then assess the appropriateness of the Internet Archive for historical web security measurements, detecting subtleties and possible pitfalls in its adoption. Finally, we investigate the feasibility of using the Internet Archive to simulate live security measurements, using recent archival data in place of live data. Our analysis shows that archive-based security measurements are a promising alternative to traditional live security measurements, which is reproducible by design; nevertheless, it also shows potential pitfalls and shortcomings of archive-based measurements. As an important contribution, we use the collected knowledge to identify insights and best practices for future archive-based security measurements.

Cross-Site Request Forgery (CSRF) is one of the oldest and simplest attacks on the Web, yet it is still effective on many websites and it can lead to severe consequences, such as economic losses and account takeovers. Unfortunately, tools and techniques proposed so far to identify CSRF vulnerabilities either need manual reviewing by human experts or assume the availability of the source code of the web application. In this paper we present Mitch, the first machine learning solution for the black-box detection of CSRF vulnerabilities. At the core of Mitch there is an automated detector of sensitive HTTP requests, i.e., requests which require protection against CSRF for security reasons. We trained the detector using supervised learning techniques on a dataset of 5,828 HTTP requests collected on popular websites, which we make available to other security researchers. Our solution outperforms existing detection heuristics proposed in the literature, allowing us to identify 35 new CSRF vulnerabilities on 20 major websites and 3 previously undetected CSRF vulnerabilities on production software already analyzed using a state-of-the-art tool.

HTTPS aims at securing communication over the Web by providing a cryptographic protection layer that ensures the confidentiality and integrity of communication and enables client/server authentication. However, HTTPS is based on the SSL/TLS protocol suites that have been shown to be vulnerable to various attacks in the years. This has required fixes and mitigations both in the servers and in the browsers, producing a complicated mixture of protocol versions and implementations in the wild, which makes it unclear which attacks are still effective on the modern Web and what is their import on web application security. In this paper, we present the first systematic quantitative evaluation of web application insecurity due to cryptographic vulnerabilities. We specify attack conditions against TLS using attack trees and we crawl the Alexa Top 10k to assess the import of these issues on page integrity, authentication credentials and web tracking. Our results show that the security of a consistent number of websites is severely harmed by cryptographic weaknesses that, in many cases, are due to external or related-domain hosts. This empirically, yet systematically demonstrates how a relatively limited number of exploitable HTTPS vulnerabilities are amplified by the complexity of the web ecosystem.

Since cookies act as the only proof of a user identity, web sessions are particularly vulnerable to session hijacking attacks, where the browser run by a given user sends requests associated to the identity of another user. When [Formula: see text] cookies are used to implement a session, there might actually be n sub-sessions running at the same website, where each cookie is used to retrieve part of the state information related to the session. Sub-session hijacking breaks the ideal view of the existence of a unique user session by selectively hijacking m sub-sessions, with [Formula: see text]. This may reduce the security of the session to the security of its weakest sub-session. In this paper, we take a systematic look at the root causes of sub-session hijacking attacks and we introduce sub-session linking as a possible defense mechanism. Out of two flavors of sub-session linking desirable for security, which we call intra-scope and inter-scope sub-session linking respectively, only the former is relatively smooth to implement. Luckily, we also identify programming practices to void the need for inter-scope sub-session linking. We finally present Warden, a server-side proxy which automatically enforces intra-scope sub-session linking on incoming HTTP(S) requests, and we evaluate it in terms of protection, performances, backward compatibility and ease of deployment.

Content Security Policy (CSP) is a recent W3C standard introduced to prevent and mitigate the impact of content injection vulnerabilities on websites. In this article, we introduce a formal semantics for the latest stable version of the standard, CSP Level 2. We then perform a systematic, large-scale analysis of the effectiveness of the current CSP deployment, using the formal semantics to substantiate our methodology and to assess the impact of the detected issues. We focus on four key aspects that affect the effectiveness of CSP: browser support, website adoption, correct configuration, and constant maintenance. Our analysis shows that browser support for CSP is largely satisfactory, with the exception of a few notable issues. However, there are several shortcomings relative to the other three aspects. CSP appears to have a rather limited deployment as yet and, more crucially, existing policies exhibit a number of weaknesses and misconfiguration errors. Moreover, content security policies are not regularly updated to ban insecure practices and remove unintended security violations. We argue that many of these problems can be fixed by better exploiting the monitoring facilities of CSP, while other issues deserve additional research, being more rooted into the CSP design.

Content Security Policy (CSP) is an emerging W3C standard introduced to mitigate the impact of content injection vulnerabilities on websites. We perform a systematic, large-scale analysis of four key aspects that impact on the effectiveness of CSP: browser support, website adoption, correct configuration and constant maintenance. While browser support is largely satisfactory, with the exception of few notable issues, our analysis unveils several shortcomings relative to the other three aspects. CSP appears to have a rather limited deployment as yet and, more crucially, existing policies exhibit a number of weaknesses and misconfiguration errors. Moreover, content security policies are not regularly updated to ban insecure practices and remove unintended security violations. We argue that many of these problems can be fixed by better exploiting the monitoring facilities of CSP, while other issues deserve additional research, being more rooted into the CSP design.

Authorization in workflow systems is usually built on top of role-based access control (RBAC), security policies on workflows are then expressed as constraints on the users performing a set of tasks and the roles assigned to them. Unfortunately, when role administration is distributed and potentially untrusted users contribute to the role assignment process, like in the case of Administrative RBAC (ARBAC), collusions may take place to circumvent the intended workflow security policies. In a collusion attack, a set of users of a workflow system collaborates by changing the user-to-role assignment, so as to sidestep the security policies and run up to completion a workflow they could not complete otherwise. In this paper, we study the problem of collusion attacks in a formal model of workflows based on stable event structures and we define a precise notion of security against collusion. We then propose a static analysis technique based on a reduction to a role reachability problem for ARBAC, which can be used to prove or disprove security for a large class of workflow systems. We also discuss how to aggressively optimise the obtained role reachability problem to ensure its tractability. Finally, we implement our analysis in a tool, WARBAC, and we experimentally show its effectiveness on a set of publicly available examples, including a realistic case study.

Model-checking is a popular approach to the security analysis of ARBAC policies, but its effectiveness is hindered by the exponential explosion of the ways in which different users can be assigned to different role combinations. In this paper we propose a paradigm shift, based on the observation that, while verifying ARBAC by exhaustive state search is complex, realistic policies often have rather simple security proofs, and we propose to use types as an effective tool to leverage this simplicity. Concretely, we present a static type system to verify the security of ARBAC policies, along with a sound and complete type inference algorithm used to automate the verification process. We then introduce compositionality results, which identify sufficient conditions to preserve the security guarantees obtained by the verification of different sub-policies when these sub-policies are combined together: this compositional reasoning is crucial when policy administration is highly distributed and naturally supports the security analysis of evolving ARBAC policies. We evaluate our approach by implementing TAPA, a static analyser for ARBAC policies based on our theory, which we test on a number of relatively large, publicly available policies from the literature.