Reference
Frameworks
Every mapping in every analysis traces to one of these primary sources. Open a framework to browse the codes, categories and layers the engine knows about.
Threat frameworks
The security taxonomies the engine maps every finding to.
OWASP Agentic Threats
T1–T17
The agentic-AI threat taxonomy from OWASP's Agentic AI Threats and Mitigations v1.1: the spine of the engine's mapping logic.
BrowseOWASP Top 10 for Agentic Applications
ASI01–ASI10
OWASP's Top 10 risks for agentic applications (ASI01–ASI10), derived per-threat from OWASP Agentic Threats cross-references via OWASP Appendix A.
BrowseMITRE ATLAS
AML.T####
MITRE's taxonomy of real-world adversary techniques against AI systems. Each report's ATLAS chips are crosswalked from the finding's OWASP Agentic Threats.
BrowseOWASP AIVSS
Core-1 – Core-10
The OWASP AIVSS Agentic AI Core Security Risks + the v0.8 numeric scoring system. Each finding gets a Core-risk category (from its ASI codes) and an advisory 0–10 AIVSS score.
BrowseSTRIDE
6 categories
Microsoft's classic threat-classification taxonomy: Spoofing, Tampering, Repudiation, Information Disclosure, DoS, Elevation of Privilege.
BrowseMAESTRO Layers
7 + Cross-Layer
The CSA 7-layer agentic-AI threat-modeling framework. The canonical layer is the universal join key across every other framework.
BrowseCompliance & AI management system
The management-system standard the mapped GCC obligations evidence.