feat(prompts): add 6 new vulnerability prompt modules

[Mr-Neutr0n]

Summary

This PR adds 6 comprehensive vulnerability prompt modules to enhance Strix agent capabilities for common but previously missing vulnerability classes.

New Modules

1. NoSQL Injection (nosql_injection.jinja)

Covers MongoDB, CouchDB, Redis, DynamoDB, and Elasticsearch with:

• Operator injection ($ne, $gt, $regex, $where)
• Authentication bypass techniques
• Blind injection (time-based, boolean-based)
• Framework-specific patterns (Mongoose, PyMongo, PHP)

2. CORS Misconfiguration (cors_misconfiguration.jinja)

Comprehensive CORS testing with:

• Origin reflection vulnerability patterns
• Null origin and subdomain exploitation
• Cache poisoning attacks
• Framework-specific vectors (Express, Flask, Django, Spring, Nginx)

3. SSTI - Server-Side Template Injection (ssti.jinja)

Coverage for 9+ template engines:

• Detection and fingerprinting payloads
• RCE chains for Jinja2, Twig, Freemarker, Velocity, Thymeleaf, Pebble, ERB, Handlebars, Smarty
• Sandbox escape techniques
• Blind SSTI exploitation

4. Prototype Pollution (prototype_pollution.jinja)

JavaScript/Node.js focused with:

• RCE gadget chains (EJS, Pug, Handlebars, child_process)
• Authentication bypass patterns
• Vulnerable npm package identification
• Framework vectors (Express, NestJS, Fastify)

5. WebSocket Security (websocket_security.jinja)

Real-time application testing:

• Cross-Site WebSocket Hijacking (CSWSH)
• Origin validation bypass
• Per-message authorization testing
• Library patterns (Socket.io, SignalR, Action Cable, GraphQL subscriptions)

6. Clickjacking (clickjacking.jinja)

UI redressing attacks with:

• X-Frame-Options and CSP frame-ancestors analysis
• Frame-buster bypass (sandbox attribute)
• OAuth authorization clickjacking
• Multi-click and cursor manipulation attacks

Module Structure

Each module follows the established format:

• Critical overview and scope definition
• Comprehensive discovery techniques
• Multiple exploitation patterns
• Bypass techniques for common defenses
• Validation steps for confirmed findings
• False positive identification
• Impact assessment
• Actionable pro tips

Testing

All modules follow the Jinja template format consistent with existing modules like csrf.jinja, idor.jinja, and sql_injection.jinja.

Checklist

• Consistent XML structure with existing modules
• Practical, actionable payloads and techniques
• Framework-specific guidance included
• Validation and false positive sections
• No sensitive/real credentials in examples

[Mr-Neutr0n]

Hi! Friendly follow-up on this PR adding 6 new vulnerability prompt modules. Let me know if there's anything I should update or if you need more context. Thanks!