Skill UI

This skill guides users through performing sophisticated web cache deception attacks. It exploits path normalization discrepancies and cache key misconfigurations between Content Delivery Networks (CDNs) and origin servers. Use this methodology to test for sensitive data exposure, cache poisoning, and authentication bypass vulnerabilities in applications protected by aggressive caching layers like Cloudflare or Akamai.

Assesses AI/ML systems for prompt injection, jailbreak, model inversion, data poisoning, and agent tool abuse using MITRE ATLAS mapping, signature detection, and scoring metrics before deployment.

A comprehensive offensive checklist for assessing the security and robustness of AI and Large Language Model (LLM) applications. It covers advanced adversarial techniques such as prompt injection, jailbreaking, model extraction, data poisoning, and analyzing system vulnerabilities across components. Essential for red-teaming and security assessment.

A comprehensive guide detailing advanced security threats in Machine Learning and AI systems. Covers model supply chain attacks (RCE via pickle, poisoning), adversarial examples (FGSM, PGD, physical-world), data privacy breaches, and LLM-specific vulnerabilities. Essential for building robust, secure AI pipelines.

A comprehensive expert playbook detailing advanced Large Language Model (LLM) prompt injection techniques. Covers direct methods (instruction override, role play), indirect attacks (RAG poisoning, web data manipulation), and sophisticated tool/function calling abuse for data exfiltration and security bypass. Essential for rigorous AI application security testing.

A comprehensive playbook detailing expert-level techniques for exploiting web cache vulnerabilities. Covers both Web Cache Deception (stealing authenticated user data) and Web Cache Poisoning (injecting malicious content). Techniques include path confusion, unkeyed header manipulation (X-Forwarded-Host), and exploiting CDN/reverse proxy caching logic. Essential reading for penetration testers and security researchers.

This skill provides a repeatable assessment for RAG systems, testing for critical vulnerabilities in vector stores and embedding models. It specifically addresses embedding inversion, cross-tenant data leakage, and knowledge-base poisoning, aligning with OWASP LLM08:2025. Use when performing authorized security audits of retrieval layers and multi-tenant isolation.

A comprehensive tool for identifying poisoned training data, backdoored ML models, and supply-chain compromises. It leverages IBM's Adversarial Robustness Toolbox (ART) for activation clustering and spectral signature analysis, uses Cleanlab for label quality checks, and enforces integrity via weight-hash verification and safetensors. Essential for securing ML pipelines before training, deployment, or during supply chain audits.

A comprehensive framework for conducting authorized security assessments on Large Language Models (LLMs) and AI agents. It covers critical vulnerabilities including advanced prompt injection (OWASP Top 10), tool abuse, RAG data poisoning, memory manipulation, and system prompt extraction. Ideal for red teaming, security auditing, and ensuring the robustness of AI systems in production environments.