AISecKG ontology as a pre-display validation layer for agent-generated content

Design · All designs
Evidence · 1 claim (1 for) · 1 study (1 design), q1 · 0 of 1 report an effect size · 1 claim rests on one study

Description

CyberAGENTS integrates the AISecKG cybersecurity ontology, which "defines valid entity types (e.g., tool, technique, attack, vulnerability, defense, system) and permissible relations among them (e.g., exploits, detects, counters, uses, can harm)." Outputs from the Challenge, Buddy, and Critic agents are checked against entity categories, relation patterns, and unsafe-content rules before learner exposure. The article reports that "Content that violates semantic constraints or safety thresholds is flagged and re-prompted under stricter conditions before presentation."

Design Implications

Context

Requirements

• Ontology-defined entity categories, relation patterns, and unsafe-content rules
• A re-prompting mechanism under stricter conditions when validation fails

Constraints

• Validation constrains cybersecurity reasoning and semantic validity; pedagogical behavior is constrained separately by schemas

Target Learners

• Undergraduate cybersecurity learners

Learning Goals

• Domain-consistent and safe instructional content in a high-stakes technical domain

Claims

• Ablating expert-informed structural components lowered LLM-judge scores on all dimensions, with largest drops in challenge quality, intent alignment, and domain grounding [+W]

Related Designs

• Structured autonomy: bounding generative agent behavior through layered pedagogical and domain constraints
• Competency-mapping module organizing curriculum topics by difficulty, prerequisites, and concept clusters

Examples

-

Key Sources

• Hornung, I., Marasinghe Arachchige, D., Kumarage, T., Agrawal, G., Deng, Y., Chen, Y.-C., & Liu, H. (2026). CyberAGENTS: Structured Autonomy for Agentic Gamified Learning in Cybersecurity. Preprint. https://arxiv.org/abs/2608.07965. doi:10.48550/arxiv.2608.07965

Page metadata

Field	Value
type	design
id	aiseckg-ontology-pre-display-validation
title	AISecKG ontology as a pre-display validation layer for agent-generated content
description	CyberAGENTS integrates the AISecKG cybersecurity ontology, which "defines valid entity types (e.g., tool, technique, attack, vulnerability, defense, system) and permissible relations among them (e.g., exploits, detect...
status	review
generated	by process:settle-candidates on 2026-10-10
trust	unverified — no one has confirmed this page's content
sources	1 — hornung-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.