TutorGPT — agentic threat model
TutorGPT presents a low-to-moderate risk profile as an educational and productivity assistant. Its primary security concerns stem from its reliance on multiple third-party LLMs and the lack of visible data privacy controls for student and educator data.
OWASP AIVSS score rationale
| Autonomy of Action | 0.20 | |
| Goal-Driven Planning | 0.30 | |
| Self-Modification | 0.10 | |
| Dynamic Tool Use | 0.40 | |
| Persistent Memory | 0.30 | |
| Contextual Awareness | 0.40 | |
| Dynamic Identity | 0.10 | |
| Multi-Agent Interactions | 0.20 | |
| Non-Determinism | 0.70 | |
| Opacity & Reflexivity | 0.60 |
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.
Integrates multiple external foundation models (GPT-5.5, Claude 4.7, Gemini 3.1, DeepSeek V4). Primary threats include prompt injection bypassing educational guardrails, model misalignment, and reliance on third-party API availability and security.
Not certain from the listing — No details are provided regarding data storage, RAG pipelines, or vector databases. General threats include potential data exfiltration of uploaded student/educator documents and lack of data lineage controls.
Not certain from the listing — The orchestration framework for the 'AI Tutor Tools' and 'task automation' is unspecified. General threats include insecure tool integration and prompt injection leading to unintended tool execution.
Not certain from the listing — Hosting, sandboxing, and API credential management details are absent. General threats include the exposure of API keys for the integrated LLM providers and lack of tenant isolation.
Not certain from the listing — No monitoring, logging, or guardrail mechanisms are described. General threats include a lack of observability into toxic, biased, or highly inaccurate educational outputs (hallucinations).
Not certain from the listing — No compliance certifications (such as COPPA/FERPA for educational tools or GDPR) or identity management controls are mentioned. General threats include unauthorized access to user accounts and data privacy violations.
Not certain from the listing — No multi-agent marketplace or agent-to-agent communication protocols are described. General threats are limited to potential future integrations of untrusted third-party educational plugins.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).