AgentReadyHomeAgent Listing

← Kwizie

Kwizie — agentic threat model

6.0AIVSS 6.0 · Medium

Kwizie is a low-risk, content-generation AI tool focused on educational quiz creation. Its primary security risks stem from processing untrusted user-uploaded files and potential prompt injection via video transcripts, rather than autonomous agentic actions.

OWASP AIVSS score rationale

AIVSS = (CVSS_Base + AARS) × Mitigation_Factor, where AARS = (10 − CVSS_Base) × (Factor_Sum / 10) × ThM
CVSS base 5.3AARS uplift 0.71Factor sum 1.6/10Threat ×0.95Mitigation ×1.0
Autonomy of Action
0.20
Goal-Driven Planning
0.10
Self-Modification
0.00
Dynamic Tool Use
0.20
Persistent Memory
0.20
Contextual Awareness
0.30
Dynamic Identity
0.00
Multi-Agent Interactions
0.00
Non-Determinism
0.40
Opacity & Reflexivity
0.20

Scored with the canonical OWASP AIVSS formula (AIVSS calculator reference); agentic risk factors estimated from the agent’s described capabilities.

MAESTRO 7-layer threat model

Per-layer threats for this agent. Layers tagged “not certain from listing” are general, caveated commentary where the public description didn’t pin that layer.

L1 · Foundation Models⚠ not certain from listing

Not certain from the listing — likely utilizes external LLMs for multilingual quiz and summary generation. Threats include prompt injection via uploaded document text or video transcripts, which could manipulate quiz outputs or bypass content filters.

L2 · Data Operations⚠ not certain from listing

Not certain from the listing — processes diverse inputs including videos (up to 3 hours), audio, text, and presentations. Threats include data exfiltration of proprietary educational content, and ingestion of malicious files designed to exploit parser vulnerabilities.

L3 · Agent Frameworks⚠ not certain from listing

Not certain from the listing — likely uses a linear pipeline rather than an autonomous agent framework. Threats are limited to insecure integration with video hosting APIs and file processing libraries.

L4 · Deployment & Infrastructure⚠ not certain from listing

Not certain from the listing — hosted as a closed-source SaaS. Threats include standard web application vulnerabilities, insecure storage of uploaded media assets, and lack of isolation during resource-intensive video processing.

L5 · Evaluation & Observability⚠ not certain from listing

Not certain from the listing — no mention of content moderation or output guardrails. Threats include the generation of inaccurate, hallucinated, or inappropriate quiz questions without detection.

L6 · Security & Compliance (cross-cutting)⚠ not certain from listing

Not certain from the listing — freemium model with certified achievements. Threats include unauthorized access to user quiz data and potential compliance gaps regarding student data privacy laws (e.g., COPPA, GDPR) if deployed in educational environments.

L7 · Agent Ecosystem⚠ not certain from listing

Not certain from the listing — does not appear to interact with external agent marketplaces or multi-agent orchestrators. Threat exposure at this layer is currently negligible.

MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).