appsec.fyi

Cross-Site Scripting (XSS) Resources

Post Share

A curated AppSec resource library covering XSS, SQLi, SSRF, IDOR, RCE, XXE, OSINT, and more.

Cross-Site Scripting (XSS)

Cross-Site Scripting (XSS) attacks are a type of injection in which malicious scripts are injected into otherwise benign and trusted websites. XSS occurs when an attacker uses a web application to send malicious code, generally in the form of a browser-side script, to a different end user.

XSS remains one of the most prevalent web vulnerabilities, appearing in everything from search bars to user profile fields. The three main variants — Reflected, Stored, and DOM-based — each have distinct attack surfaces. Reflected XSS executes via a crafted URL, Stored XSS persists in the application's database and fires for every visitor, and DOM-based XSS exploits client-side JavaScript that unsafely handles user input without any server round-trip.

The impact of XSS extends well beyond simple alert boxes. Attackers leverage it for session hijacking, credential theft, keylogging, phishing overlays, and as a pivot point for deeper exploitation. In bug bounty programs, Stored XSS on authenticated pages consistently pays well because it can be chained into account takeover.

Modern defenses include Content Security Policy (CSP), output encoding, and frameworks that auto-escape by default — but bypasses are discovered regularly, making XSS a constantly evolving attack surface.

This page collects research, bypass techniques, payloads, and real-world writeups covering all forms of cross-site scripting.

From OWASP

Start here: the XSS learning pathAn ordered route through this library — orientation, then methodology, then research. Read the XSS guideA long-form, source-cited deep dive synthesized from every resource below. The comprehensive XSS guide on chs.usA hand-written, in-depth practitioner guide — attacks, testing, and prevention.
Date Added Link Excerpt
2026-08-26 NEW 2026The WAF Blocked My XSS — So I Rotated What It Was Reading intermediateThis post from InfoSec Write-ups details a security researcher's experience bypassing a Web Application Firewall (WAF) that was blocking their XSS attempts. The core of their strategy involved "rotating what it was reading," implying they found a way to alter or obfuscate their malicious input such that the WAF could no longer detect it, allowing the XSS exploit to succeed. The post promises to elaborate on this technique. No bounty amount is mentioned. → infosecwriteups.com
2026-08-26 NEW 2026What's in a tag name? JavaScript, apparently beginner 3 min readWriteup detailing JavaScript XSS vulnerabilities exploitable through manipulated HTML tag names. The research explores how browsers' lenient parsing of tag names and properties like `localName`, `part`, and `classList` can be leveraged to construct effective attack vectors. Specific techniques discussed include using `tabindex`, `contenteditable`, `onfocus` event chaining, the `Function` constructor, and bypassing WAFs by incorporating malformed characters and unusual attribute manipulations. → portswigger.net
2026-08-17 2026DOM Purify - dirty namespace bypass intermediate 2 min read AuthZWriteup on a DOMPurify SVG sanitization bypass that exploits a flaw in the `DATA_ATTR` regular expression. By omitting an end-of-string anchor, an attacker can inject a colon into a data attribute name, allowing for custom namespaces. This technique enables the creation of a `data-namespace:attribute` like `data-slonser:href="javascript:alert(1)"` within SVG files when `PARSER_MEDIA_TYPE` is set to `application/xhtml+xml`, leading to XSS execution. The vulnerability was reported and fixed by adding the missing anchor to the regex.
2026-08-17 2026Make Self-XSS Great Again intermediate 7 min readTechnique for transforming Self-XSS into Stored XSS by leveraging credentialless iframes. This method allows attackers to bypass session limitations, enabling attacks like credential theft and account takeover by exploiting vulnerable login forms with CSRF or clickjacking, and even managing CAPTCHAs through WebSocket communication.
2026-08-14 2026[tl;dr sec] #341 - Hugging Face Incident Black Hat Talk, CSS Bomb in your Inbox, GitHub Supply Chain Security Improvements news 10 min read Supply Chain TalksReference detailing incidents and advancements in application security, including a deep dive into the OpenAI-Hugging Face incident presented at Black Hat, exploitation of CSS within webmail clients for token exfiltration and UI spoofing, and GitHub's security improvements for npm and Actions to disrupt supply chain attacks. It also covers an AI-discovered pre-authentication RCE chain in WordPress core and a CDN caching vulnerability in RubyGems.org. → tldrsec.com
2026-08-09 2026CVE-2026-64638: Critical Pre-Auth XSS Vulnerability in WordPress Allows Remote Code Execution news 4 min readWriteup of CVE-2026-64638, a critical pre-authentication reflected XSS vulnerability in WordPress versions up to 7.0.2. This flaw, exploited via a technique dubbed XSS2Shell, allows unauthenticated attackers to execute JavaScript on the login page, potentially leading to PHP code execution and full site compromise. Escalation to remote code execution requires tricking an administrator into visiting a malicious page, enabling the creation of application passwords and plugin uploads. Immediate patching to WordPress 7.0.3 is crucial. → rescana.com
2026-08-09 2026High-Severity WordPress Vulnerability Affects All Versions Could Lead to Admin Account Takeover news 4 min readWriteup of CVE-2026-64638, a pre-authentication reflected XSS on the WordPress login page, enabling attackers to achieve PHP code execution via an "XSS2Shell" attack chain that leverages Application Passwords and Same Origin Method Execution (SOME). This vulnerability, discovered by pwn.ai, affects all WordPress versions and can lead to admin account takeover, website hijacking, and arbitrary code execution. WordPress has released emergency patches.
2026-08-07 2026PimpMyCaido #1: Hunt client-side vulnerabilities with DOMLogger++ intermediate MobileThis content introduces PimpMyCaido #1, a guide focused on finding client-side vulnerabilities. It specifically highlights the use of a tool called DOMLogger++ for this purpose. The title suggests a practical, hands-on approach to security testing within web applications, emphasizing techniques for uncovering weaknesses accessible from the user's browser. → yeswehack.com
2026-08-07 2026CSS:the bomb inside your inbox intermediate 37 min readAnalysis of CSS sanitization bypass techniques, demonstrating novel methods to exfiltrate tokens, compromise third-party websites, and steal passwords by exploiting discrepancies between browser rendering and sanitizer interpretation. Techniques discussed include abusing HTML `<label>` elements for UI control in Outlook and employing `:before` and `:after` pseudo-elements with opacity manipulation to achieve indirect prompt injection in OpenAI's Atlas browser, targeting webmail clients like Yahoo Mail, AOL Mail, Fastmail, ProtonMail, GMail, and Outlook. → portswigger.net
2026-08-06 2026Stored XSS in Django's admin via an unvalidated URLField display path (CVE-2026-15920) intermediate 8 min readWriteup of CVE-2026-15920, detailing a stored XSS vulnerability in Django's admin. This flaw occurs when URLField values are displayed as clickable links without proper validation, allowing attacker-controlled `javascript:` URIs to execute arbitrary code within an authenticated staff session. The vulnerability stems from a coding oversight where the `FileField` branch's structure was copied for `URLField` without retaining necessary scheme checks. This bypasses standard validation, as many write paths do not invoke `full_clean()`, making it possible to inject malicious data directly into the database.
2026-08-02 2026Russian hackers exploit unpatched Zimbra servers to steal emails news 2 min readWriteup of CVE-2025-66376, a cross-site scripting vulnerability in Zimbra Collaboration Suite exploited by Russian hacker group Laundry Bear to steal emails and sensitive data from government and commercial networks. The attack weaponizes specially crafted HTML emails, requiring only viewing to execute JavaScript, enabling the theft of account data, authentication tokens, and application passcodes. Laundry Bear has also been observed using AI in the development of their infrastructure for data exfiltration. → helpnetsecurity.com
2026-07-28 2026How I Found a Bug Worth $3,500 — In a Feature Nobody Was Watching. intermediate 6 min read Bug Bounty RCEWriteup detailing a $3,500 bounty discovery involving a file upload form on a B2B SaaS platform. The exploit chained a CWE-434 and CWE-770 storage exhaustion flaw, achievable by spoofing file size metadata, with a CWE-79 critical stored XSS in the filename that executed in an admin's browser. The analysis highlights the importance of scrutinizing metadata, understanding victim context for severity, testing "boring" features, and chaining vulnerabilities. → infosecwriteups.com
2026-07-23 2026Finding eight high-severity vulnerabilities in NodeBB in six hours intermediate 19 min read AuthZ Bug BountyWriteup detailing eight high-severity vulnerabilities discovered in NodeBB versions prior to 4.14.0, including Cross-Site Scripting (XSS) exploiting custom Federation servers and template injection. The writeup specifies a template injection impacting nearly all inputs and authorization bypasses allowing data hijacking. The analysis highlights how autonomous agents achieved these findings within hours, leading to prompt fixes by NodeBB maintainers. → aikido.dev
2026-07-18 2026SAR 2,629 For Stored XSS via svg Image Leading to ATO intermediate 2 min readWriteup detailing a Stored XSS vulnerability found via profile picture uploads, enabling Account Takeover (ATO). Exploiting a missing Content Security Policy (CSP) and same-origin hosting of SVG images allowed for the execution of injected JavaScript. This script accessed and exfiltrated authentication tokens stored in `localStorage`, granting full control of the user account without needing credentials. Remediation involves SVG sanitization, strict CSP implementation, and enforcing `Content-Disposition: attachment`. → infosecwriteups.com
2026-07-08 2026Chaining a DOM XSS Sink, WAF Bypass, Cross-Origin Smuggling, and SDK Abuse into One Click Account… advanced 8 min read API Sec AuthNLibrary for chaining DOM XSS, WAF bypass via `window.name` cross-origin smuggling, and SDK abuse to achieve one-click account takeover. The technique exploits a `javascript:` URL sink, bypasses Akamai's WAF by inserting characters between keywords and parentheses, leverages `window.name` persistence across navigations, and abuses a first-party authentication SDK to retrieve signed JWTs and live AWS STS credentials. → infosecwriteups.com
2026-06-29 2026CVE-2026-13536: Reflected XSS Vulnerability in GotoHTTP Remote Access Platform (reg.12x Endpoint) Analysis and Mitigation intermediate 3 min readAnalysis of CVE-2026-13536 details a reflected XSS vulnerability in the GotoHTTP remote access platform's /reg.12x endpoint, stemming from improper sanitization of the sn parameter. This allows unauthenticated attackers to inject JavaScript via crafted URLs, with a public PoC available on GitHub. While no active exploitation by APT groups is reported, and it's not on the CISA KEV catalog, opportunistic attacks are a risk due to the low complexity and user interaction requirement. The vendor has acknowledged the issue and will remove the vulnerable parameter in their next release. → rescana.com
2026-06-25 2026CVE-2026-10086: High-Severity XSS Vulnerability in GitLab Enterprise Edition Analytics Dashboard Analysis Impact and Mitigation Steps news 3 min readAnalysis of CVE-2026-10086, a high-severity XSS vulnerability in GitLab Enterprise Edition's Analytics Dashboard, details how authenticated attackers can inject JavaScript. This vulnerability, CWE-79, allows for session hijacking and privilege escalation by executing code in user contexts. Patched versions include 19.1.1, 19.0.3, and 18.11.6. While not yet observed in the wild, prompt patching and log review are advised, with potential mitigation by restricting dashboard access. → rescana.com
2026-06-23 2026CVE-2026-25860 turn XSS to RCE intermediateCVE-2026-25860 turn XSS to RCE
2026-06-23 2026Exploring WebExtension security vulnerabilities in React Developer Tools and Vue.js devtools intermediate 5 min read Bug BountyWriteup detailing WebExtension security vulnerabilities, including unverified external messages in React Developer Tools (CVE-2023-5654) allowing arbitrary URL fetching and unauthorized access to page capture APIs in Vue.js devtools (CVE-2023-5718) leading to screenshot data leakage. This research highlights risks inherent in the WebExtension architecture and its components, affecting cross-browser compatibility and user data. → snyk.io
2026-06-22 2026Exploiting Auth0 Defaults in XSS Attacks - elttam intermediate 8 min read AuthNWriteup detailing how XSS vulnerabilities in applications using Auth0 can be exploited. The article highlights the insecure implicit grant flow, enabled by default in Auth0, and demonstrates how it can be combined with other misconfigurations to pivot across tenant applications. Specifically, it shows how an attacker can leverage XSS to steal access tokens intended for a protected API, facilitating lateral movement within an Auth0 tenant. → elttam.com
2026-06-21 2026Understanding and mitigating the Jinja2 XSS vulnerability (CVE-2024-22195) intermediate 3 min read PythonReference detailing CVE-2024-22195, a cross-site scripting vulnerability in Jinja2 versions prior to 3.1.3. The vulnerability arises from the `xmlattr` filter when processing user input with spaces in keys, allowing attackers to inject arbitrary HTML attributes and potentially execute untrusted scripts. Mitigation involves upgrading to Jinja2 3.1.3 and utilizing tools like Snyk for continuous monitoring and detection of vulnerable dependencies in Python projects and Docker containers. → snyk.io
2026-06-20 2026“Bug Bounty Bootcamp #48: OAuth + XSS ” intermediate AuthN Bug BountyThis "Bug Bounty Bootcamp #48" article, titled "OAuth + XSS," explores a potent combination of vulnerabilities: OAuth and Cross-Site Scripting (XSS). The content suggests that by leveraging these two, attackers can achieve account takeovers, effectively describing it as an "ultimate account takeover one-two punch." The article is part of a series and can be found on InfoSec Write-ups. No specific bounty payout amount is mentioned. → infosecwriteups.com
2026-06-17 2026How to prevent log injection vulnerability in JavaScript and Node.js applications intermediate 6 min readLibrary for preventing log injection vulnerabilities in JavaScript and Node.js applications, specifically detailing how attackers can manipulate input to inject malicious code into logs. It offers methods for sanitizing user inputs, using regex and libraries like validator.js, suggests careful consideration of what data to log, and recommends structured logging and specialized libraries such as pino over basic console.log. The entry also mentions the Snyk IDE extension for VS Code as a tool for detecting such vulnerabilities. → snyk.io
2026-06-16 2026Automatically fix code vulnerabilities with AI intermediate 4 min read AILibrary for automatically fixing common security vulnerabilities, such as Cross-site Scripting (XSS) in Java applications, by leveraging a hybrid AI model. This tool, integrated into IDEs, goes beyond providing remediation advice by directly applying secure code fixes, exemplified in a Spring Boot application using the Thymeleaf template engine and the faker library. Unlike generative AI assistants that may introduce insecure code, this library uses a combination of generative AI, symbolic AI, and machine learning, trained on curated security research data, to ensure secure code generation. → snyk.io
2026-06-14 2026MeshCentral: From XSS to RCE intermediate 8 min read RCEWriteup detailing the exploitation of MeshCentral, moving from Cross-Site Scripting (XSS) to Remote Code Execution (RCE). The author demonstrates how an LLM, Claude Opus, was utilized to identify vulnerabilities and generate proof-of-concept exploits by analyzing the MeshCentral agent and server interactions. The process involved extracting agent credentials from local files to impersonate existing agents and ultimately achieve RCE, showcasing a practical application of AI in vulnerability research.
2026-06-13 2026Six levels, one lesson: LLMs cannot keep a secret intermediate AIGitHub's Secure Code Game Season 3, a six-level challenge, demonstrates that Large Language Models (LLMs) cannot keep secrets. The game involves prompt injection attacks against vulnerable AI assistants designed to hide information. Players craft attacks to extract these secrets, highlighting that system prompts are not a security measure. The content focuses on hands-on learning of AI security principles through this open-source, browser-based game. No bounty payout amounts are mentioned. → infosecwriteups.com
2026-06-12 2026Chaining Stored XSS and CSRF in Typemill CMS: A Deep Dive into Attribute Injection intermediate CSRFA security assessment of Typemill CMS uncovered a critical vulnerability chain combining Stored XSS and CSRF (CVE-2026–53468). An attacker can bypass frontend validation to inject malicious scripts into page metadata. This allows for the theft of admin sessions by exploiting attribute injection. This vulnerability impacts Typemill, a popular flat-file CMS built on PHP and the Slim framework, and poses a significant risk to its users. → infosecwriteups.com
2026-06-08 2026JavaScript Prototype Pollution Deep Dive : — Reconnaissance, Exploitation & Bug Bounty Guideline advanced RCEThis article provides a deep dive into JavaScript Prototype Pollution vulnerabilities, explaining the underlying prototype chain and its attack vectors. It covers reconnaissance methodologies, exploitation techniques ranging from XSS to Remote Code Execution (RCE), and real-world bug bounty case studies. The guide also delves into advanced exploit chains, tooling, automation, and defense strategies, offering a production-ready Python scanner. The content focuses on understanding and mitigating this complex JavaScript vulnerability. → infosecwriteups.com
2026-06-08 2026From XSS to RCE (dompdf 0day) intermediate 10 min read RCELibrary for Remote Code Execution (RCE) in dompdf, a popular PHP library used for rendering PDFs from HTML. The vulnerability, identified as a 0-day by Positive Security, allows an attacker to inject CSS that tricks dompdf into caching a malicious font file with a `.php` extension. This file can then be executed remotely by accessing it from the web server. The exploit leverages the `$isRemoteEnabled` setting and the font caching mechanism within dompdf.
2026-06-03 2026https://github.com/Armur-Ai/Pentest-Swarm-AI beginner 6 min read AI ReconLibrary for advanced penetration testing utilizing a real swarm intelligence architecture. It coordinates independent agents via stigmergy and emergence, allowing them to coordinate by writing to and reading from a shared blackboard, rather than through a central planner. This approach enables emergent attack chains and dynamic agent interaction, supporting tools like nmap, sqlmap, Burp, ZAP, and Metasploit, and is compatible with LLMs such as Claude and Llama.
2026-05-28 2026CVE-2026-41241: Critical Stored XSS in Pretalx Conference Platform Allows Attackers 100% Talk Acceptance (Patched in 2026.1.0) news 5 min readWriteup of CVE-2026-41241, a critical stored XSS vulnerability in Pretalx versions prior to 2026.1.0, allowing any registered user to compromise organizer accounts and force talk acceptance. Exploitation involves submitting a talk proposal with a crafted XSS payload in fields like title, speaker display name, or email, which executes when an organizer uses the backend search. The vulnerability stems from improper sanitization and unsafe `innerHTML` usage. Immediate upgrade to version 2026.1.0 is recommended. → rescana.com
2026-05-20 2026CVE-2026-42897 Zero-Day Analysis: Microsoft Exchange Server OWA XSS Vulnerability Exploited in the Wild news 5 min readAnalysis of CVE-2026-42897 details a zero-day cross-site scripting (XSS) vulnerability affecting on-premises Microsoft Exchange Server 2016, 2019, and Subscription Edition. Actively exploited in the wild, this flaw in Outlook Web Access (OWA) allows attackers to execute arbitrary JavaScript, leading to session hijacking and credential theft. The analysis covers threat actor TTPs, exploitation evidence, and actionable mitigations like the Exchange Emergency Mitigation Service (EEMS) and the Exchange On-premises Mitigation Tool (EOMT), noting potential side effects such as the loss of OWA Print Calendar functionality. → rescana.com
2026-05-17 2026Unpatched Microsoft Exchange Server vulnerability exploited (CVE-2026-42897) newsMicrosoft Exchange Server is vulnerable to exploitation due to an unpatched security flaw, identified as CVE-2026-42897. Attackers can leverage this vulnerability, impacting systems that have not been updated. This poses a significant risk to organizations using Microsoft Exchange Server. Further details on the exploitation and its potential impact can be found via the provided link. → helpnetsecurity.com
2026-05-12 2026Instructure confirms hackers used Canvas flaw to deface portals news 2 min readWriteup on ShinyHunters exploiting cross-site scripting (XSS) vulnerabilities in Instructure's Canvas LMS. Attackers used these flaws to gain authenticated admin sessions, deface login portals with extortion messages, and exfiltrate over 3.6 terabytes of data. The attacks targeted the Free-for-Teacher environment, leading to temporary downtime and account closures. → bleepingcomputer.com
2026-05-09 2026Every Old Vulnerability Is Now an AI Vulnerability beginnerThis article argues that as Artificial Intelligence (AI) systems become more integrated, traditional cybersecurity vulnerabilities are now also AI vulnerabilities. Existing exploits and weaknesses in software, hardware, and network infrastructure can be leveraged to target or compromise AI models. This means that the vast landscape of known security flaws presents a significant risk to AI systems, requiring a re-evaluation of security strategies to account for this expanded threat surface. → darkreading.com
2026-05-03 2026'Chaining vulnerabilities is the hallmark of a sophisticated attack': 750000 websites must be patched as Microsoft's popular open source Dotnetnuke CMS hit by an XSS flaw that allows attackers to hijack admin sessions and take over entire web servers news 2 min readLibrary for securing DotNetNuke CMS, addressing CVE-2026-40321, a cross-site scripting (XSS) flaw. This vulnerability allows attackers to upload malicious SVG files, which, when clicked by an authenticated administrator, execute JavaScript, hijack sessions, and enable arbitrary file writes to the server via the `/API/personaBar/ConfigConsole/UpdateConfigFile` endpoint. This enables the creation of ASPX web shells for full server compromise, impacting over 750,000 websites built on the Microsoft-backed platform.
2026-04-30 2026dr34mhacks/XSSNow: Find XSS payloads that actually work by filtering them based on real-world constraints instead of blind payload spraying. intermediate 2 min readLibrary of curated XSS payloads, XSSNow aids researchers and bug bounty hunters by providing context-aware, defense-focused, and real-world tested payloads. It categorizes vulnerabilities by injection context, offers specific collections for WAF bypasses and encoding evasions, and suggests payloads optimized for character limitations and filters. The platform also details CSP bypass techniques and browser quirks, encouraging community contributions to its knowledge base of HTML injection, attribute breaking, JavaScript context, CSS injection, and URL parameter attacks.
2026-04-24 2026Over 10000 Zimbra servers vulnerable to ongoing XSS attacks news 2 min readWriteup of CVE-2025-48700, an ongoing XSS vulnerability impacting over 10,000 Zimbra Collaboration Suite instances. Exploitable by unauthenticated attackers, this flaw allows arbitrary JavaScript execution, enabling sensitive information access. Patched in June 2025, it has been actively abused in the wild, leading to CISA's inclusion in its Known Exploited Vulnerabilities Catalog and an order for Federal Civilian Executive Branch agencies to secure affected servers. Previous Zimbra vulnerabilities have also been exploited by APT28 and Russian Winter Vivern. → bleepingcomputer.com
2026-04-22 2026Mass-Assignment to Stored XSS and CSP Bypass in a Chatbot Platform advancedMass-Assignment to Stored XSS and CSP Bypass in a Chatbot Platform
2026-04-22 2026Full Disclosure: DOM-Based XSS And Failures In Bug Bounty Hunting beginner 8 min readWriteup detailing a DOM-based XSS vulnerability discovered in a bug bounty hunt, where an insecure `eUrl` parameter on a login page allowed for dynamic resource loading from an attacker-controlled server. This flaw, combined with the absence of the HTTPOnly flag on the `ASPSESSIONID` cookie, enabled a one-click account takeover. The writeup emphasizes the importance of input validation, sanitization, allow-listing, CSP, and proper cookie flag implementation to mitigate such risks.
2026-04-22 2026Cross-Site Scripting (XSS) Is Surging: 4 New CVEs This Week news 1 min readWriteup of surging Cross-Site Scripting (XSS) vulnerabilities, detailing four new CVEs including CVE-2026-27243 with a CVSS score of 9.3. It highlights the increasing prevalence of XSS in SaaS environments, the limitations of automated scanners, and the need for regular testing of controls like WAFs and EDRs against current attack patterns, referencing n8n webhooks abused for malware delivery.
2026-04-22 2026CVE-2025-26244: Stored XSS in DeimosC2 Leading to Privilege Escalation news 7 min readWriteup detailing CVE-2025-26244, a stored cross-site scripting vulnerability in CyberOneSecurity's DeimosC2 v1.1.0-Beta. The writeup demonstrates how an attacker can register a malicious agent by reverse-engineering an agent binary to obtain listener details. This allows injection of an XSS payload into the 'graph' endpoint, which executes when a user views the graph. The stolen cookie then enables privilege escalation and unauthorized access to the C2 framework.
2026-04-22 2026CVE-2025-25461: SeedDMS Stored XSS newsWriteup of CVE-2025-25461, a Stored XSS vulnerability in SeedDMS 6.0.29. Exploitable by users with "Add Category" permissions, an attacker can inject XSS payloads into category names, leading to execution when documents associated with that category are viewed. Potential impacts include session hijacking, data exfiltration, phishing, and remote code execution. Mitigation involves sanitizing user input, employing CSP, and proper output encoding.
2026-04-22 2026Finding DOM Polyglot XSS in PayPal the Easy Way intermediate 6 min readLibrary for discovering DOM-based polyglot XSS vulnerabilities. It details a process utilizing Burp Suite's embedded browser and DOM Invader to identify insecure sinks, specifically on PayPal. The library also demonstrates how to bypass Content Security Policy (CSP) by exploiting unintended script gadgets within the PayPal application, including leveraging older versions of Bootstrap and a custom `youtube.js` gadget to execute JavaScript. → portswigger.net
2026-04-22 2026Cisco IOS XE Web Authentication Reflected XSS Advisory newsCisco IOS XE Web Authentication Reflected XSS Advisory
2026-04-22 2026CVE-2025-66412: Angular Stored XSS via SVG Animation and MathML Attributes news 2 min readWriteup of CVE-2025-66412, an Angular Stored XSS vulnerability. The Angular Template Compiler's incomplete security schema allows bypassing sanitization for URL-holding attributes and SVG animation elements. Attackers can inject `javascript:` URLs into attributes like `xlink:href` or by manipulating the `attributeName` in SVG animations, leading to arbitrary code execution, session hijacking, and data exfiltration. Patches are available in Angular versions 19.2.17, 20.3.15, and 21.0.2.
2026-04-22 2026CVE-2025-0133: PAN-OS Reflected XSS in GlobalProtect Gateway news 2 min readWriteup detailing CVE-2025-0133, a reflected XSS vulnerability in Palo Alto Networks PAN-OS GlobalProtect gateway and portal. This flaw allows attackers to execute malicious JavaScript in a user's browser via crafted links, primarily posing a risk of phishing and credential theft, especially when Clientless VPN is enabled. Mitigation involves enabling Threat IDs 510003 and 510004 via Threat Prevention content version 8995, applying Vulnerability Protection profiles to security rules, or disabling Clientless VPN.
2026-04-22 2026Synthesizing XSS Polyglots with Monte Carlo Tree Search (arXiv 2025) advancedSynthesizing XSS Polyglots with Monte Carlo Tree Search (arXiv 2025) → arxiv.org
2026-04-19 2026Bypassing Signature-Based XSS Filters: Modifying HTML intermediate 2 min readTechnique for bypassing signature-based XSS filters by modifying HTML syntax, demonstrating methods to obfuscate payloads. It explores variations in tag casing, insertion of NULL bytes and superfluous characters, use of alternative attribute delimiters like backticks, and HTML encoding within attribute values to evade detection. Examples are provided using DVWA and OWASP's Broken Web Application Project. → portswigger.net
2026-04-19 2026Advanced XSS Filter Bypass Methods Using Payload Splitting advancedAdvanced XSS Filter Bypass Methods Using Payload Splitting

Frequently Asked Questions

What are the three types of XSS?
The three main types are Reflected XSS (payload delivered via a URL and immediately reflected in the response), Stored XSS (payload persisted in the application database and served to other users), and DOM-based XSS (payload executed entirely in the browser via client-side JavaScript without a server round-trip).
How do you prevent cross-site scripting?
Key defenses include output encoding (HTML, JavaScript, URL, and CSS contexts), Content Security Policy (CSP) headers, using frameworks that auto-escape by default (React, Angular), input validation, and the HttpOnly flag on session cookies to limit the impact of successful attacks.
Why is XSS still so common?
XSS persists because web applications have many injection points (URL parameters, form fields, headers, file uploads), developers must encode output correctly for every context, and modern JavaScript frameworks can be bypassed through dangerouslySetInnerHTML, template injection, or prototype pollution.

Weekly AppSec Digest

Get new resources delivered every Monday.