← All Terms

AI Vulnerability Scoring System

AIVSS

An OWASP framework that adds an agentic uplift to a CVSS v4.0 score, based on ten agent capabilities such as autonomy, tools and persistent memory.

Agentic AI

The AI Vulnerability Scoring System is an OWASP project for scoring flaws in agentic AI. Version 0.8 takes a CVSS v4.0 base score as the floor, then scores ten agent factors at 0, 0.5 or 1: autonomy, tools, natural language control, context, non-determinism, opacity, persistent memory, identity, multi-agent interaction and self-modification. Their sum sets how much of the gap to 10 is added, adjusted for exploit maturity and mitigation strength.

The idea behind it is that an agent multiplies the damage of an ordinary flaw. A SQL injection in a reporting tool is serious; the same flaw in an agent that can read the results and email them out is far worse, though the code defect is identical.

AIVSS is still before version 1.0, and its authors call its weights provisional. Its most useful instruction is procedural: rank agent findings on the adjusted score, read scores by severity band, and expect the order to differ from the CVSS order.

Read more on this
Security

Agent-Aware Vulnerability Management Ranks Flaws by What the Agent Can Do

A vulnerability score describes a flaw. It says nothing about the AI agent standing behind it: how freely it acts, what it is allowed to touch, which tools it can call, what it remembers and which business process it runs. Agent-aware vulnerability management adds those five things to exploitability before deciding what to fix first. ISO/IEC 27090:2026, published in October, gives the threat side of the same picture. Here is how the two fit together and how to run it.