Skip to main content

Cyberstrike is now open source! AI-powered penetration testing for security professionals. Star on GitHub

Web Application Agent

The Web Application agent is a specialized security testing agent equipped with comprehensive knowledge of the OWASP Web Security Testing Guide (WSTG) v4.2, featuring 109 documented tests across 12 categories.

Overview

The Web Application agent combines deep security knowledge with automated tooling to perform thorough web application penetration testing. It’s pre-loaded with specialized skills for SQL injection testing and OWASP WSTG methodology.

Agent Configuration

PropertyValue
Agent IDweb-application
ColorRed
Pre-loaded Skillsowasp-wstg, sql-injection
Allowed ToolsBrowser, Bash, WebFetch, WebSearch, Read, Glob, Grep

Launch the Agent

Terminal window
# Start Cyberstrike with web application agent
cyberstrike --agent web-application
# Or switch during session
/agent web-application

Knowledge Base

The Web Application agent has access to extensive security knowledge through the skills system.

OWASP WSTG v4.2 Skill

The agent is pre-loaded with the complete OWASP Web Security Testing Guide v4.2, providing systematic testing methodology for 109 security tests across 12 categories.

SQL Injection Skill

Comprehensive SQL injection testing knowledge including:

  • Detection payloads (basic, error-based)
  • Database fingerprinting queries (MySQL, PostgreSQL, MSSQL, Oracle, SQLite)
  • Union-based injection methodology
  • Blind SQL injection (boolean-based, time-based)
  • Out-of-band (OOB) techniques with DNS exfiltration
  • WAF bypass techniques
  • SQLMap command reference

OWASP WSTG v4.2 Coverage

The agent systematically tests all WSTG categories:

WSTG-INFO: Information Gathering (10 tests)

Test IDTest Name
WSTG-INFO-01Conduct Search Engine Discovery Reconnaissance
WSTG-INFO-02Fingerprint Web Server
WSTG-INFO-03Review Webserver Metafiles for Information Leakage
WSTG-INFO-04Enumerate Applications on Webserver
WSTG-INFO-05Review Webpage Content for Information Leakage
WSTG-INFO-06Identify Application Entry Points
WSTG-INFO-07Map Execution Paths Through Application
WSTG-INFO-08Fingerprint Web Application Framework
WSTG-INFO-09Fingerprint Web Application
WSTG-INFO-10Map Application Architecture

WSTG-CONF: Configuration Testing (13 tests)

Test IDTest Name
WSTG-CONF-01Test Network Infrastructure Configuration
WSTG-CONF-02Test Application Platform Configuration
WSTG-CONF-03Test File Extensions Handling
WSTG-CONF-04Review Old Backup and Unreferenced Files
WSTG-CONF-05Enumerate Infrastructure and Admin Interfaces
WSTG-CONF-06Test HTTP Methods
WSTG-CONF-07Test HTTP Strict Transport Security
WSTG-CONF-08Test RIA Cross Domain Policy
WSTG-CONF-09Test File Permission
WSTG-CONF-10Test for Subdomain Takeover
WSTG-CONF-11Test Cloud Storage
WSTG-CONF-12Test for Content Security Policy
WSTG-CONF-13Test for Path Confusion

Tip

The agent automatically tests for S3 bucket misconfigurations and Azure Blob storage exposure under WSTG-CONF-11.

WSTG-IDNT: Identity Management (5 tests)

Test IDTest Name
WSTG-IDNT-01Test Role Definitions
WSTG-IDNT-02Test User Registration Process
WSTG-IDNT-03Test Account Provisioning Process
WSTG-IDNT-04Test for Account Enumeration
WSTG-IDNT-05Test for Weak or Unenforced Username Policy

WSTG-ATHN: Authentication Testing (11 tests)

Test IDTest Name
WSTG-ATHN-01Test for Credentials Transported over Encrypted Channel
WSTG-ATHN-02Test for Default Credentials
WSTG-ATHN-03Test for Weak Lock Out Mechanism
WSTG-ATHN-04Test for Bypassing Authentication Schema
WSTG-ATHN-05Test for Vulnerable Remember Password
WSTG-ATHN-06Test for Browser Cache Weaknesses
WSTG-ATHN-07Test for Weak Password Policy
WSTG-ATHN-08Test for Weak Security Question Answer
WSTG-ATHN-09Test for Weak Password Change or Reset
WSTG-ATHN-10Test for Weaker Authentication in Alternative Channel
WSTG-ATHN-11Test Multi-Factor Authentication

WSTG-ATHZ: Authorization Testing (7 tests)

Test IDTest Name
WSTG-ATHZ-01Test Directory Traversal File Include
WSTG-ATHZ-02Test for Bypassing Authorization Schema
WSTG-ATHZ-03Test for Privilege Escalation
WSTG-ATHZ-04Test for Insecure Direct Object References (IDOR)
WSTG-ATHZ-05Test for OAuth Weaknesses
WSTG-ATHZ-06Test for JWT Weaknesses
WSTG-ATHZ-07Test for GraphQL Weaknesses

WSTG-SESS: Session Management (11 tests)

Test IDTest Name
WSTG-SESS-01Test for Session Management Schema
WSTG-SESS-02Test for Cookie Attributes
WSTG-SESS-03Test for Session Fixation
WSTG-SESS-04Test for Exposed Session Variables
WSTG-SESS-05Test for Cross Site Request Forgery (CSRF)
WSTG-SESS-06Test for Logout Functionality
WSTG-SESS-07Test Session Timeout
WSTG-SESS-08Test for Session Puzzling
WSTG-SESS-09Test for Session Hijacking
WSTG-SESS-10Test JSON Web Tokens
WSTG-SESS-11Test for Cookie Scope

WSTG-INPV: Input Validation Testing (19 tests)

Test IDTest Name
WSTG-INPV-01Test for Reflected Cross Site Scripting
WSTG-INPV-02Test for Stored Cross Site Scripting
WSTG-INPV-03Test for HTTP Verb Tampering
WSTG-INPV-04Test for HTTP Parameter Pollution
WSTG-INPV-05Test for SQL Injection
WSTG-INPV-06Test for LDAP Injection
WSTG-INPV-07Test for XML Injection
WSTG-INPV-08Test for SSI Injection
WSTG-INPV-09Test for XPath Injection
WSTG-INPV-10Test for IMAP SMTP Injection
WSTG-INPV-11Test for Code Injection
WSTG-INPV-12Test for Command Injection
WSTG-INPV-13Test for Format String Injection
WSTG-INPV-14Test for Incubated Vulnerability
WSTG-INPV-15Test for HTTP Splitting Smuggling
WSTG-INPV-16Test for HTTP Incoming Requests
WSTG-INPV-17Test for Host Header Injection
WSTG-INPV-18Test for Server-Side Template Injection (SSTI)
WSTG-INPV-19Test for Server-Side Request Forgery (SSRF)

Danger

SQL Injection (WSTG-INPV-05) testing uses the pre-loaded sql-injection skill with comprehensive payloads for all major database systems.

WSTG-ERRH: Error Handling (2 tests)

Test IDTest Name
WSTG-ERRH-01Test for Improper Error Handling
WSTG-ERRH-02Test for Stack Traces

WSTG-CRYP: Cryptography (4 tests)

Test IDTest Name
WSTG-CRYP-01Test for Weak Transport Layer Security
WSTG-CRYP-02Test for Padding Oracle
WSTG-CRYP-03Test for Sensitive Information Sent via Unencrypted Channels
WSTG-CRYP-04Test for Weak Encryption

WSTG-BUSL: Business Logic (10 tests)

Test IDTest Name
WSTG-BUSL-01Test Business Logic Data Validation
WSTG-BUSL-02Test Ability to Forge Requests
WSTG-BUSL-03Test Integrity Checks
WSTG-BUSL-04Test for Process Timing
WSTG-BUSL-05Test Number of Times a Function Can Be Used Limits
WSTG-BUSL-06Test for Circumvention of Work Flows
WSTG-BUSL-07Test Defenses Against Application Misuse
WSTG-BUSL-08Test Upload of Unexpected File Types
WSTG-BUSL-09Test Upload of Malicious Files
WSTG-BUSL-10Test Payment Functionality

WSTG-CLNT: Client-Side Testing (14 tests)

Test IDTest Name
WSTG-CLNT-01Test for DOM-Based Cross Site Scripting
WSTG-CLNT-02Test for JavaScript Execution
WSTG-CLNT-03Test for HTML Injection
WSTG-CLNT-04Test for Client-Side URL Redirect
WSTG-CLNT-05Test for CSS Injection
WSTG-CLNT-06Test for Client-Side Resource Manipulation
WSTG-CLNT-07Test Cross Origin Resource Sharing
WSTG-CLNT-08Test for Cross Site Flashing
WSTG-CLNT-09Test for Clickjacking
WSTG-CLNT-10Test WebSockets
WSTG-CLNT-11Test Web Messaging
WSTG-CLNT-12Test Browser Storage
WSTG-CLNT-13Test for Cross Site Script Inclusion
WSTG-CLNT-14Test for Reverse Tabnabbing

WSTG-APIT: API Testing (4 tests)

Test IDTest Name
WSTG-APIT-01Test for GraphQL
WSTG-APIT-02Test for REST API
WSTG-APIT-03Test for SOAP API
WSTG-APIT-04Test for WebSocket API

Testing Methodology

The agent follows a structured approach:

┌─────────────────────────────────────────────────────────────┐
│ TESTING METHODOLOGY │
├─────────────────────────────────────────────────────────────┤
│ 1. RECONNAISSANCE │
│ └── Information gathering, fingerprinting │
│ │
│ 2. MAPPING │
│ └── Application structure, entry points │
│ │
│ 3. DISCOVERY │
│ └── Vulnerability identification per WSTG category │
│ │
│ 4. EXPLOITATION │
│ └── Proof of concept, impact demonstration │
│ │
│ 5. REPORTING │
│ └── Findings with CVSS scores, remediation │
└─────────────────────────────────────────────────────────────┘

Integrated Security Tools

The agent leverages industry-standard tools:

ToolPurpose
NucleiTemplate-based vulnerability scanning
SQLMapAutomated SQL injection exploitation
FFUFWeb fuzzing and content discovery
NiktoWeb server vulnerability scanner
curlHTTP request crafting and testing

Tool Integration Example

> Use sqlmap to test the login endpoint for SQL injection
[Executing SQLMap...]
sqlmap -u "https://target.com/api/login" --data="username=test&password=test" --batch --level=5 --risk=3
[SQLMap Output]
Parameter: username (POST)
Type: boolean-based blind
Payload: username=test' AND 5765=5765 AND 'test'='test
Type: time-based blind
Payload: username=test' AND SLEEP(5) AND 'test'='test
Database: MySQL >= 5.0.12

Browser-Based Testing

The agent uses the HackerBrowser for dynamic analysis with full traffic capture:

> Launch browser and navigate to https://target.com/login
> Test the username field for XSS vulnerabilities
[Browser launched]
[Navigating to /login...]
[Injecting XSS payloads...]
Testing payload: <script>alert('XSS')</script>
Testing payload: <img src=x onerror=alert('XSS')>
Testing payload: javascript:alert('XSS')
Testing payload: <svg onload=alert('XSS')>
FINDING: Reflected XSS
━━━━━━━━━━━━━━━━━━━━━━━
WSTG-ID: WSTG-INPV-01
Location: GET /login?error=<message>
Parameter: error
Payload: <svg onload=alert(document.domain)>
Severity: High (CVSS 6.1)
Evidence: JavaScript executed in browser context
CWE: CWE-79

CVSS Severity Rating

The agent rates findings using CVSS 3.1:

SeverityCVSS ScoreExample Vulnerabilities
Critical9.0 - 10.0RCE, Auth Bypass, SQLi with data exfil
High7.0 - 8.9Stored XSS, IDOR with sensitive data
Medium4.0 - 6.9Reflected XSS, CSRF, Info disclosure
Low0.1 - 3.9Missing headers, verbose errors

Session Example

You: Test the e-commerce site at shop.example.com for OWASP Top 10 vulnerabilities
Agent: I'll perform comprehensive security testing following OWASP WSTG methodology.
## Phase 1: Information Gathering
[Fingerprinting web server...]
Server: nginx/1.18.0
Framework: React (frontend), Node.js/Express (backend)
Technologies: JWT authentication, PostgreSQL database
[Mapping application structure...]
Found 47 endpoints across 8 API routes
## Phase 2: Authentication Testing (WSTG-ATHN)
[Testing WSTG-ATHN-02: Default Credentials...]
✓ No default credentials found
[Testing WSTG-ATHN-04: Auth Bypass...]
⚠ FINDING: JWT Algorithm Confusion
FINDING: JWT Algorithm Confusion
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
WSTG-ID: WSTG-SESS-10
Location: Authorization header
Parameter: JWT token
Payload: Algorithm changed from RS256 to HS256
Severity: Critical (CVSS 9.8)
Evidence: Forged admin token accepted
CWE: CWE-347
## Phase 3: Authorization Testing (WSTG-ATHZ)
[Testing WSTG-ATHZ-04: IDOR...]
⚠ FINDING: Insecure Direct Object Reference
FINDING: IDOR in Order History
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
WSTG-ID: WSTG-ATHZ-04
Location: GET /api/orders/{id}
Parameter: id
Payload: Sequential ID enumeration
Severity: High (CVSS 7.5)
Evidence: Accessed orders of other users
CWE: CWE-639
[Continuing with remaining WSTG categories...]

Best Practices

  1. Scope Definition - Clearly define testing boundaries and excluded endpoints
  2. Authentication - Provide valid credentials for comprehensive authenticated testing
  3. Rate Limiting - Configure appropriate delays to avoid triggering WAF/rate limits
  4. Evidence Collection - Use HAR export for complete traffic capture
  5. Verification - Manually verify critical and high severity findings

Caution

Always obtain proper written authorization before testing web applications. Unauthorized security testing is illegal and unethical.