Tag: Web Application Firewall

  • The API Battlefield: Akamai’s 2026 Report Unmasks the Staggering 113% Surge in Interface Attacks

    Akamai has promulgated its annual State of the Internet dossier, chronicling the landscape of applications, APIs, and distributed denial-of-service (DDoS) bombardments, and has subsequently chronicled a profound metamorphosis in adversarial stratagems. The cardinal revelation distills to this singular truth: kinetic strikes have evolved into architectures of profound systemic complexity, become precipitously more economical to scale, and are now inextricably entwined with the very infrastructure through which enterprises proliferate digital services and integrate artificial intelligence. APIs now reside at the epicenter of this mounting pressure. Until recently, a multitude of enterprises relegated them to the periphery of their defensive perimeters; presently, however, APIs are increasingly weaponized as the paramount vector of ingress.

    Forensic investigators observe a marked departure from isolated, clamorous campaigns orchestrated merely for notoriety and reputational reverberations. Vastly more prevalent is the architecting of kinetic strikes as meticulously synchronized operations, seamlessly orchestrating the subjugation of APIs, assaults upon web applications, and Layer 7 (Application Layer) DDoS bombardments—striking at the very heart of the OSI model. This multifaceted paradigm empowers digital marauders not merely to shatter service availability, but to exponentially inflate the victim’s infrastructural expenditures. As enterprises plunge deeper into the crucible of AI and digital automation, adversaries exhibit a heightened appetite for striking the very interfaces and services that sustain these architectures.

    The telemetry enshrined within this nascent dossier unequivocally demonstrates that we are no longer witnessing isolated paroxysms. Over the preceding biennium, the volume of Layer 7 DDoS bombardments has surged by a staggering 104%. Between 2023 and 2025, the frequency of kinetic strikes directed at web applications skyrocketed by 73%. Furthermore, the daily average of API assaults experienced an astronomical 113% year-over-year escalation. Akamai concurrently illuminates the findings of an enterprise survey: a chilling 87% of respondents conceded enduring at least one API-centric security breach during 2025. This aggregation of telemetry unequivocally proves that APIs have long transcended the realm of esoteric technical discourse amongst developers, crystallizing into a paramount defensive frontline.

    In Akamai’s estimation, the very teleology of these attacks is undergoing a profound transfiguration. Malefactors increasingly strive not merely to breach the citadel and exfiltrate telemetry, but to actively degrade service efficacy, paralyze applications, precipitate a hemorrhagic surge in computational resource expenditures, and aggressively co-opt AI automation to serve their own nefarious machinations. This paradigm is profoundly advantageous for the assailants on several fronts. Foremost, automation drastically diminishes the fiscal burden of preparation. Secondly, pre-fabricated operational scripts facilitate the rapid replication of identical kinetic maneuvers across disparate quarries. Thirdly, bombardments directed at APIs and web applications frequently yield a substantial financial harvest, even bereft of a classic, headline-grabbing penetration.

    The dossier emphatically underscores an auxiliary tribulation: the bifurcation of application security and API security is, in practical reality, an untenable paradigm. A multitude of enterprises persist in managing these domains as entirely disparate endeavors, wielding divergent instruments, mobilizing isolated squadrons, and suffering from fragmented visibility. Such a labyrinthine architecture inevitably breeds chasms in oversight. For the digital marauder, these very blind spots manifest as highly fortuitous vectors of ingress, given that within the crucible of an authentic attack, the web application and the API are customarily weaponized as a singular, unified conduit.

    The document further harbors a multitude of auxiliary observations that eloquently illustrate the shifting trajectory of peril. One such observation is tethered to the phenomenon colloquially christened “vibe coding”—a paradigm wherein code is synthesized at breakneck velocity, frequently leaning heavily upon AI auxiliaries, whilst utterly forsaking rigorous engineering discipline. In Akamai’s estimation, this methodology increasingly injects nascent vulnerabilities and configurational aberrations directly into the production environment, entirely bypassing the crucible of rigorous pre-deployment validation. Distilled to its essence: enterprises are simultaneously hyper-accelerating their developmental cadence whilst precipitously eroding their safety margins, and adversaries are lying in wait, eager to exploit these hastily deployed interfaces within the operational theater.

    A dedicated sector of the dossier is devoted to the scrutiny of DDoS kinetic activity orchestrated by hacktivist syndicates. Akamai chronicles that politically motivated cabals are relentlessly amplifying their pressure, fueled by the volatile permutations of the international geopolitical tableau and the burgeoning accessibility of rentable botnet architectures. This contemporary infrastructure bears little resemblance to the rudimentary, artisanal networks of subjugated devices characteristic of bygone eras. The digital bazaar is increasingly dominated by the “DDoS-for-hire” and “DDoS-as-a-Service” (DDoSaaS) paradigms, wherein requisite destructive capacity is procured merely as a commodity. As the friction to access such arsenals diminishes, the threshold for entry plummets, inviting a deluge of nascent participants.

    The researchers unequivocally tether the 104% explosion in Layer 7 bombardments directly to this profound accessibility. It has become terrifyingly simple for malefactors to procure a bespoke botnet and seamlessly augment it with AI-fortified attack scripts. Consequently, the meticulous curation of targets is simplified, the fiscal burden of the operation plummets, and the ignition of campaigns against APIs and web applications is hyper-accelerated. The dossier specifically illuminates the specter of “super-botnets” akin to Aisuru and Kimwolf. These leviathan networks are cultivating architectures that trace their lineage to the infamous Mirai contagion, presently serving as the foundational bedrock for the DDoS-as-a-Service ecosystem. Crucially, this infrastructure is patronized not solely by cybercriminal syndicates, but equally by fervent hacktivists.

    Akamai concurrently directs its gaze toward the overarching economic macro-environment. The contemporary internet bombardment is increasingly architected as a ruthless business model, wherein supreme efficiency reigns paramount. Whereas historically, a digital marauder was compelled to expend colossal resources upon labyrinthine, manual preparation, contemporary paradigms allow for the automation of myriad tasks, whilst bespoke instruments can simply be leased. This evolution renders the kinetic strike not merely scalable, but predictably, effortlessly repeatable. For the vanguard of defenders, this metamorphosis is exceptionally harrowing; the theater of conflict is no longer defined by sparse, complex operations, but rather by an unrelenting torrent of economical, hyper-accelerated campaigns capable of being ignited ad infinitum.

    This nascent dossier transcends mere macroeconomic statistics, offering a granular dissection of regional trajectories, an appraisal of the economic engines driving contemporary internet bombardments, and a bespoke editorial by a guest luminary regarding the fortification against nascent perils tethered to agentic AI architectures. Within this context, “agentic AI” customarily denotes systems that do not merely regurgitate responses to inquiries, but possess the autonomy to execute labyrinthine chains of action, invoke auxiliary instruments, and seamlessly commune with external services. For the defensive vanguard, this paradigm is acutely perilous; an agent relies almost exclusively upon APIs, meaning that a singular vulnerability or architectural aberration within the interface instantaneously cascades, compromising the entirety of the broader automation chain.

    The State of the Internet series has endured as a beacon of insight for its twelfth consecutive year. Akamai traditionally anchors its deductions upon the colossal telemetry harvested through its sovereign, global defensive infrastructure—an architecture that processes a monumental fraction of planetary web traffic. Within the contemporary iteration of this document, the paramount focus has irrevocably shifted toward the labyrinthine nexus of applications, APIs, DDoS bombardments, and artificial intelligence. In essence, the dossier chronicles a brutally stark reality: whilst enterprises hyper-accelerate their digital transfigurations, digital marauders possess the agility to adapt to these nascent architectures with virtually zero latency. And if APIs have indeed become the bedrock of AI services, then the pragmatic defense of artificial intelligence inextricably begins with the impregnable fortification of the API.

  • WAFSmith: A New Open-Source Tool Uses LLMs to Revolutionize WAF Management

    Leveraging on LLM’s abilities to mimic cognitive human agents, WAFSmith aims to reduce the friction of WAF rule governance from rule creation to deployment in minutes. It is designed as a highly disruptive tool to augment Blue Team operations in a rapidly evolving threat landscape. It was developed to enhance Blue Team’s capabilities to respond to threats in a fast and effective manner, without compromising business operations. The solution is first of the kind, especially in the open source landscape, a novel approach to solve a challenging problem of WAF rule governance.

    Use

    [pastacode lang=”markup” manual=”wafsmith%20-h%0A%0AUsage%3A%20wafsmith%20%5Boptions%5D%20%5Bcommand%5D%0A%0AWAFSmith%3A%20LLM%20based%20WAF%20Rule%20Creation%20Framework%0A%0AOptions%3A%0A%20%20-V%2C%20–version%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20output%20the%20version%20number%0A%20%20-h%2C%20–help%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20display%20help%20for%20command%0A%0ACommands%3A%0A%20%20create%20%5Boptions%5D%20%3Cinput%3E%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20Create%20ModSecurity%20Rules%0A%20%20evaluate%20%5Boptions%5D%20%3Cpayload-directory%3E%20%20Evaluate%20deployed%20WAF%20rules%20against%20a%20list%20of%20payloads%0A%20%20extract%20%5Boptions%5D%20%3Cinput%3E%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20Extract%20payload(s)%20from%20logs%0A%20%20aggregate%20%5Boptions%5D%20%3Cinput%3E%20%20%20%20%20%20%20%20%20%20%20%20%20Optimizing%20ModSecurity%20Rules%0A%20%20help%20%5Bcommand%5D%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20display%20help%20for%20command” message=”” highlight=”” provider=”manual”/]

    extract

    [pastacode lang=”markup” manual=”wafsmith%20extract%20-h%0A%0AUsage%3A%20wafsmith%20extract%20%5Boptions%5D%20%3Cinput%3E%0A%0AExtract%20payload(s)%20from%20logs%0A%0AArguments%3A%0A%20%20input%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20Input%20directory%20%2F%20file%20containing%20the%20Nginx%20log%20files%0A%0AOptions%3A%0A%20%20-o%2C%20–output%20%3Coutput-file%3E%20%20Specify%20the%20output%20file%20for%20the%20newly%20generated%20rule(s)%20if%20any%0A%20%20-k%2C%20–api-key%20%3Ckey%3E%20%20%20%20%20%20%20%20%20OpenAI%20API%20Key%0A%20%20-b%2C%20–base-url%20%3Cbase%3E%20%20%20%20%20%20%20OpenAI%20SDK%20Endpoint%0A%20%20-l%2C%20–model%20%3Cmodel%3E%20%20%20%20%20%20%20%20%20OpenAI%20Model%0A%20%20-x%2C%20–threads%20%3Cthreads%3E%20%20%20%20%20Specify%20the%20number%20of%20threads%20to%20be%20used%20during%20the%20rule%20generation%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20process.%20Default%20is%2010%20(default%3A%2010)%0A%20%20-h%2C%20–help%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20display%20help%20for%20command” message=”” highlight=”” provider=”manual”/]

    evaluate

    [pastacode lang=”markup” manual=”wafsmith%20evaluate%20-h%0A%0AUsage%3A%20wafsmith%20evaluate%20%5Boptions%5D%20%3Cpayload-directory%3E%0A%0AEvaluate%20deployed%20WAF%20rules%20against%20a%20list%20of%20payloads%0A%0AArguments%3A%0A%20%20payload-directory%20%20%20%20%20%20%20%20%20%20%20%20%20%20Input%20directory%20%2F%20file%20containing%20the%20payloads%0A%0AOptions%3A%0A%20%20-e%2C%20–evaded%20%3Cevaded-file%3E%20%20%20%20%20Specify%20the%20output%20file%20for%20evaded%20payload(s)%20if%20any%0A%20%20-s%2C%20–setup%20%3Csetup-directory%3E%20%20Specify%20the%20directory%20which%20contains%20the%20docker%20compose%20enviornment%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20setup%0A%20%20-t%2C%20–traffic%20%3Ctraffic%3E%20%20%20%20%20%20%20%20Specify%20directory%20%2F%20file%20containing%20business%20traffic%20content%20for%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20simulation%0A%20%20-p%2C%20–position%20%3Cposition%3E%20%20%20%20%20%20Specify%20the%20postion%20of%20the%20payload%20in%20the%20HTTP%20request.%20Default%20is%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20url-parameters%20(default%3A%20%22url-parameters%22)%0A%20%20-m%2C%20–method%20%3Cmethod%3E%20%20%20%20%20%20%20%20%20%20Specify%20the%20HTTP%20method%20for%20the%20payload.%20Default%20is%20GET.%20(default%3A%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%22GET%22)%0A%20%20-x%2C%20–threads%20%3Cthreads%3E%20%20%20%20%20%20%20%20Specify%20the%20number%20of%20threads%20to%20be%20used%20during%20the%20rule%20generation%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20process.%20Default%20is%2010%20(default%3A%2010)%0A%20%20-h%2C%20–help%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20display%20help%20for%20command” message=”” highlight=”” provider=”manual”/]

    create

    [pastacode lang=”markup” manual=”wafsmith%20create%20-h%0A%0AUsage%3A%20wafsmith%20create%20%5Boptions%5D%20%3Cinput%3E%0A%0ACreate%20ModSecurity%20Rules%0A%0AArguments%3A%0A%20%20input%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20Input%20directory%20%2F%20file%20containing%20the%20payloads%0A%0AOptions%3A%0A%20%20-o%2C%20–output%20%3Coutput-file%3E%20%20%20%20%20Specify%20the%20output%20file%20for%20the%20newly%20generated%20rule(s)%20if%20any%0A%20%20-e%2C%20–evaded%20%3Cevaded-file%3E%20%20%20%20%20Specify%20the%20output%20file%20for%20evaded%20payload(s)%20if%20any%0A%20%20-s%2C%20–setup%20%3Csetup-directory%3E%20%20Specify%20the%20directory%20which%20contains%20the%20docker%20compose%20enviornment%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20setup%0A%20%20-t%2C%20–traffic%20%3Ctraffic%3E%20%20%20%20%20%20%20%20Specify%20directory%20%2F%20file%20containing%20business%20traffic%20content%20for%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20simulation%0A%20%20-k%2C%20–api-key%20%3Ckey%3E%20%20%20%20%20%20%20%20%20%20%20%20OpenAI%20API%20Key%0A%20%20-b%2C%20–base-url%20%3Cbase%3E%20%20%20%20%20%20%20%20%20%20OpenAI%20SDK%20Endpoint%0A%20%20-l%2C%20–model%20%3Cmodel%3E%20%20%20%20%20%20%20%20%20%20%20%20OpenAI%20Model%0A%20%20-p%2C%20–position%20%3Cposition%3E%20%20%20%20%20%20Specify%20the%20postion%20of%20the%20payload%20in%20the%20HTTP%20request.%20Default%20is%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20url-parameters%20(default%3A%20%22url-parameters%22)%0A%20%20-m%2C%20–method%20%3Cmethod%3E%20%20%20%20%20%20%20%20%20%20Specify%20the%20HTTP%20method%20for%20the%20payload.%20Default%20is%20GET.%20(default%3A%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%22GET%22)%0A%20%20-x%2C%20–threads%20%3Cthreads%3E%20%20%20%20%20%20%20%20Specify%20the%20number%20of%20threads%20to%20be%20used%20during%20the%20rule%20generation%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20process.%20Default%20is%2010%20(default%3A%2010)%0A%20%20-h%2C%20–help%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20display%20help%20for%20command%0A” message=”” highlight=”” provider=”manual”/]

    aggregate

    [pastacode lang=”markup” manual=”wafsmith%20aggregate%20-h%0A%0AUsage%3A%20wafsmith%20aggregate%20%5Boptions%5D%20%3Cinput%3E%0A%0AOptimizing%20ModSecurity%20Rules%0A%0AArguments%3A%0A%20%20input%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20Input%20directory%20%2F%20file%20containing%20the%20list%20of%20payloads%20that%20the%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20rules%20will%20are%20designed%20to%20catch%0A%0AOptions%3A%0A%20%20-r%2C%20–rules%20%3Crules%3E%20%20%20%20%20%20%20%20%20%20%20%20Input%20directory%20%2F%20file%20containing%20the%20list%20of%20rules%20that%20will%20be%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20aggregated%0A%20%20-o%2C%20–output%20%3Coutput-file%3E%20%20%20%20%20Specify%20the%20output%20file%20for%20the%20newly%20generated%20rule(s)%20if%20any%0A%20%20-e%2C%20–evaded%20%3Cevaded-file%3E%20%20%20%20%20Specify%20the%20output%20file%20for%20evaded%20payload(s)%20if%20any%0A%20%20-s%2C%20–setup%20%3Csetup-directory%3E%20%20Specify%20the%20directory%20which%20contains%20the%20docker%20compose%20enviornment%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20setup%0A%20%20-t%2C%20–traffic%20%3Ctraffic%3E%20%20%20%20%20%20%20%20Specify%20directory%20%2F%20file%20containing%20business%20traffic%20content%20for%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20simulation%0A%20%20-k%2C%20–api-key%20%3Ckey%3E%20%20%20%20%20%20%20%20%20%20%20%20OpenAI%20API%20Key%0A%20%20-b%2C%20–base-url%20%3Cbase%3E%20%20%20%20%20%20%20%20%20%20OpenAI%20SDK%20Endpoint%0A%20%20-l%2C%20–model%20%3Cmodel%3E%20%20%20%20%20%20%20%20%20%20%20%20OpenAI%20Model%0A%20%20-p%2C%20–position%20%3Cposition%3E%20%20%20%20%20%20Specify%20the%20postion%20of%20the%20payload%20in%20the%20HTTP%20request.%20Default%20is%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20url-parameters%20(default%3A%20%22url-parameters%22)%0A%20%20-m%2C%20–method%20%3Cmethod%3E%20%20%20%20%20%20%20%20%20%20Specify%20the%20HTTP%20method%20for%20the%20payload.%20Default%20is%20GET.%20(default%3A%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%22GET%22)%0A%20%20-x%2C%20–threads%20%3Cthreads%3E%20%20%20%20%20%20%20%20Specify%20the%20number%20of%20threads%20to%20be%20used%20during%20the%20rule%20generation%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20process.%20Default%20is%2010%20(default%3A%2010)%0A%20%20-h%2C%20–help%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20display%20help%20for%20command” message=”” highlight=”” provider=”manual”/]

    Install

  • The WAF Deception: 70% of Firewalls Bypassed by HTTP Parameter Pollution and JS Injection

    A recent automated study conducted by ETHIACK has revealed that modern web application security mechanisms—including widely adopted Web Application Firewalls (WAFs)—are vulnerable to a novel class of attacks that combine JavaScript injection with HTTP parameter pollution. Tests encompassing products from major cloud platforms and cybersecurity vendors demonstrated that over 70% of WAF configurations could be bypassed using carefully crafted requests.

    The key to this attack lies in the discrepancies between how web applications and protective systems process duplicate parameters in HTTP requests. The vulnerability was first identified in an ASP.NET application protected by a strictly configured WAF. ASP.NET has the characteristic behavior of merging multiple parameters with the same name into a single value, separated by commas. This behavior becomes critical when such parameters are interpreted within a JavaScript context—even seemingly innocuous values can be transformed into executable code.

    For instance, a request such as /​?q=1’&q=alert(1)&q=’2 results in the server generating the string 1′,alert(1),’2, which, when embedded into a JavaScript line, becomes syntactically valid and leads to code execution. The culprit is the comma operator in JavaScript, which allows sequential execution of expressions. This opens a path for injecting malicious scripts while evading the signature-based filters of most WAFs.

    The technique, known as HTTP Parameter Pollution, is not new. However, its fusion with JavaScript injection significantly enhances its effectiveness. Researchers tested 17 different WAF configurations from AWS, Google Cloud, Microsoft Azure, Cloudflare, and others. Simple payloads bypassed about 17% of defenses, but when parameter pollution was applied, the success rate surged to 70%.

    Only five configurations successfully withstood all test cases: a specific ruleset from Azure WAF, Google Cloud Armor, and three setups from open-appsec. Particularly vulnerable were three AWS WAF rulesets, each of which failed every test, allowing all malicious payloads to pass through unimpeded. The broader trend highlighted a clear advantage of machine learning–based systems over traditional signature-based ones: behavioral models proved far more adept at detecting subtly obfuscated attack patterns.

    Nonetheless, even advanced solutions harbored weaknesses. The researchers’ autonomous “hackbot” discovered a misconfigured Azure WAF instance that could be circumvented using a rudimentary payload: test\’;alert(1);//. This underscores the reality that even cloud-native security solutions, often marketed as industry-leading, remain susceptible to elementary evasion techniques.

    The researchers emphasize that a WAF cannot function as a standalone security perimeter in the presence of insecure application code. The existence of such vulnerabilities points to a fundamental disconnect between input processing logic in protective systems and web applications. Without architectural re-evaluation and robust input sanitization within the application itself, even the most expensive WAFs cannot guarantee effective protection against parsing-based attack vectors.

  • UUSEC WAF: Free, Industrial-Grade Web Application Firewall with AI & Three-Layer Defense

    UUSEC WAF Web Application Firewall is an industrial grade free, high-performance, and highly scalable web application and API security protection product that supports AI and semantic engines. It is a comprehensive website protection product launched by UUSEC Technology, which first realizes the three-layer defense function of traffic layer, system layer, and runtime layer.

    Web Application Firewall, AI Security

     Technical advantages

    Intelligent 0-day defense

    UUSEC WAF innovatively applies machine learning technology, using anomaly detection algorithms to distinguish and identify HTTP normal and attack traffic, and models whitelist threats to normal traffic. By using machine learning algorithms to automatically learn the parameter characteristics of normal traffic and convert them into corresponding parameter whitelist rule libraries, it is possible to intercept attacks without adding rules when facing various sudden 0-day vulnerabilities, eliminating the pain of website managers having to work late to upgrade as soon as vulnerabilities appear.

    Ultimate CDN acceleration

    UUSEC self-developed cache cleaning feature surpasses the arbitrary cache cleaning function only available in the commercial version of nginx, proxy_cache_purge. The commercial version of nginx only supports * pattern matching to clean the cache, while UUSEC WAF further supports regular expression matching URL path cache cleaning, which has higher flexibility and practicality compared to the commercial version of nginx. Users can enjoy ultimate CDN acceleration while more conveniently solving cache expiration issues.

    Powerful proactive defense

    The self-developed ‘HIPS’ and ‘RASP’ functions of UUSEC WAF can achieve more powerful dual layer defense at the system layer and application runtime layer, effectively preventing zero day vulnerability attacks. Host layer active defense can intercept low-level attacks at the system kernel layer, such as restricting process network communication, process creation, file read and write, system privilege escalation, system overflow attacks, etc. Runtime application self-defense RASP is inserted into runtime engines such as Java JVM and PHP Zend to effectively track runtime context and intercept various web 0-day vulnerability attacks.

    Advanced semantic engine

    UUSEC WAF adopts four industry-leading semantic analysis based detection engines, namely SQL, XSS, RCE, and LFI. Combined with multiple deep decoding engines, it can truly restore HTTP content such as base64, JSON, and form data, effectively resisting various attack methods that bypass WAF. Compared with traditional regular matching, it has the characteristics of high accuracy, low false alarm rate, and high efficiency. Administrators do not need to maintain a complex rule library to intercept multiple types of attacks.

    Advanced rule engine

    UUSEC WAF actively utilizes the high-performance and highly flexible features of nginx and luajit. In addition to providing a traditional rule creation mode that is user-friendly for ordinary users, it also offers a highly scalable and flexible Lua script rule writing function, allowing advanced security administrators with certain programming skills to create a series of advanced vulnerability protection rules that traditional WAF cannot achieve. Users can write a series of plugins to extend the existing functions of WAF. This makes it easier to intercept complex vulnerabilities.

    Install & Use

  • Web Application Firewall (WAF) Comparison Project

    Web Application Firewall (WAF) Comparison Project

    This project repository contains testing datasets and tools to compare WAF efficacy in the two most important categories:

    • Security Coverage (True Positive Rate) – measures the WAF’s ability to correctly identify and block malicious requests is crucial in today’s threat landscape. It must preemptively block zero-day attacks as well as effectively tackle known attack techniques utilized by hackers
    • Precision (False Positive Rate) – measures the WAF’s ability to correctly allow legitimate requests. Any hindrance to these valid requests could lead to significant business disruption and an increased workload for administrators.

    This project aims to measure the efficacy of each WAF against a variety of legitimate and malicious HTTP requests, taken from real-world scenarios.

    The project is described in detail in this blog.

    Methodology

    Each WAF solution is tested against two data sets: legitimate and malicious. We then used a formula described below in detail to produce a single balanced score.

    Legitimate Data Set

    The Legitimate Requests Dataset is carefully designed to test WAF behaviors in real-world scenarios. To attain this, it includes 973,964 different HTTP requests from 185 real-web sites in 12 categories. Each dataset was recorded by browsing to real-world websites and conducting various operations in the site (for example, sign-up, selecting products and placing in a cart, etc) ensuring the presence of 100% legitimate requests.

    The dataset can be found in the folder Data/Legitimate

    Malicious Data Set

    The Malicious Requests Dataset includes 73,924 malicious payloads from a broad spectrum of commonly experienced attack vectors:

    • SQL Injection-
    • Cross-Site Scripting (XSS)
    • XML External Entity (XXE)
    • Path Traversal
    • Command Execution
    • Log4Shell
    • Shellshock

    The malicious payloads were sourced from the WAF Payload Collection GitHub page that was assembled by mgm security partners GmbH from Germany. This repository serves as a valuable resource, providing payloads specifically created for testing Web Application Firewall rules.

    The dataset is available here

    Tooling

    To trigger the data sets through the different devices under test, we developed a simple test tool in Python. The test tool is designed to ingest data sets as input and send each request to the various WAFs being tested. It reads the data files from the data sets and uses the requests module in a multi-threaded manner to send the data to each WAF.

    During the initial phase, the tool conducts a dual-layer health check for each WAF. This process first validates connectivity to each WAF, ensuring system communication. It then checks that each WAF is set to prevention mode, confirming its ability to actively block malicious requests.

    The responses from each request sent by the test tool to the WAFs were systematically logged in a dedicated database for further analysis. The database we used is an AWS RDS instance running PostgreSQL (the database is not included in this repo). You can configure it to work with any SQL database of your preference by adjusting the settings in the config.py file.

    Install & Use

    Copyright (C) 2023 openappsec

  • coraza: OWASP Coraza Web Application Firewall

    OWASP Coraza Web Application Firewall

    Welcome to OWASP Coraza WAF, Coraza is a golang enterprise-grade Web Application Firewall framework that supports Modsecurity’s seclang language and is 100% compatible with OWASP Core Ruleset.

     

    Web Application Firewall

    Coraza v2 differences with v1

    • Full internal API refactor, public API has not changed
    • Full audit engine refactor with plugins support
    • New enhanced plugins interface for transformations, actions, body processors, and operators
    • We are fully compliant with Seclang from modsecurity v2
    • Many features were removed and transformed into plugins: XML (Mostly), GeoIP, and PCRE regex
    • Better debug logging
    • New error logging (like modsecurity)

    Why Coraza WAF?

    Philosophy

    • Simplicity: Anyone should be able to understand and modify Coraza WAF’s source code
    • Extensibility: It should be easy to extend Coraza WAF with new functionalities
    • Innovation: Coraza WAF isn’t just a ModSecurity port. It must include awesome new functions (in the meantime, it’s just a port ?)
    • Community: Coraza WAF is a community project, and all ideas will be considered

    Install & Use

    Copyright 2021 Juan Pablo Tosso

  • BunkerWeb: Open-source and next-generation Web Application Firewall (WAF)

    BunkerWeb

    BunkerWeb is a next-generation and open-source Web Application Firewall (WAF).

    Being a full-featured web server (based on NGINX under the hood), it will protect your web services to make them “secure by default”. BunkerWeb integrates seamlessly into your existing environments (LinuxDockerSwarmKubernetes, …) and is fully configurable (don’t panic, there is an awesome web UI if you don’t like the CLI) to meet your own use-cases . In other words, cybersecurity is no more a hassle.

    Why BunkerWeb?

    • Easy integration into existing environments : Seamlessly integrate BunkerWeb into various environments such as Linux, Docker, Swarm, Kubernetes and more. Enjoy a smooth transition and hassle-free implementation.
    • Highly customizable : Tailor BunkerWeb to your specific requirements with ease. Enable, disable, and configure features effortlessly, allowing you to customize the security settings according to your unique use case.
    • Secure by default : BunkerWeb provides out-of-the-box, hassle-free minimal security for your web services. Experience peace of mind and enhanced protection right from the start.
    • Awesome web UI : Take control of BunkerWeb more efficiently with the exceptional web user interface (UI). Navigate settings and configurations effortlessly through a user-friendly graphical interface, eliminating the need for the command-line interface (CLI).
    • Plugin system : Extend the capabilities of BunkerWeb to meet your own use cases. Seamlessly integrate additional security measures and customize the functionality of BunkerWeb according to your specific requirements.
    • Free as in “freedom” : BunkerWeb is licensed under the free AGPLv3 license, embracing the principles of freedom and openness. Enjoy the freedom to use, modify, and distribute the software, backed by a supportive community.
    • Professional services : Get technical support, tailored consulting and custom development directly from the maintainers of BunkerWeb. Visit the Bunker Panel for more information.

    Security features

    A non-exhaustive list of security features :

    • HTTPS support with transparent Let’s Encrypt automation
    • State-of-the-art web security : HTTP security headers, prevent leaks, TLS hardening, …
    • Integrated ModSecurity WAF with the OWASP Core Rule Set
    • Automatic ban of strange behaviors based on HTTP status code
    • Apply connections and requests limit for clients
    • Block bots by asking them to solve a challenge (e.g. : cookie, javascript, captcha, hCaptcha or reCAPTCHA)
    • Block known bad IPs with external blacklists and DNSBL
    • And much more …

    Install & Use