Aitherapy — agentic threat model
Aitherapy presents low agentic risk due to its lack of tool execution and autonomous planning, but poses high data privacy and safety risks due to the handling of sensitive mental health conversations without explicit compliance or clinical guardrails.
OWASP AIVSS score rationale
| Autonomy of Action | 0.10 | |
| Goal-Driven Planning | 0.10 | |
| Self-Modification | 0.00 | |
| Dynamic Tool Use | 0.00 | |
| Persistent Memory | 0.20 | |
| Contextual Awareness | 0.40 | |
| Dynamic Identity | 0.00 | |
| Multi-Agent Interactions | 0.00 | |
| Non-Determinism | 0.60 | |
| Opacity & Reflexivity | 0.50 |
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.
Not certain from the listing — likely uses a commercial or open-source LLM fine-tuned or prompted for therapeutic conversation. Vulnerable to jailbreaks, prompt injection, and generating harmful or misaligned medical/psychological advice.
Not certain from the listing — likely stores chat histories and user profiles. Vulnerable to data exfiltration of sensitive mental health conversations and lack of HIPAA-compliant data handling or encryption at rest.
Not certain from the listing — likely a simple conversational wrapper rather than a complex agentic framework. Vulnerable to session state manipulation or prompt injection bypassing therapeutic guardrails.
Not certain from the listing — hosted as a web application (aitherapy.care). Vulnerable to standard web application threats, lack of secure sandboxing for user sessions, and potential data leaks via insecure APIs.
Not certain from the listing — no mention of automated guardrails, clinical evaluation, or real-time toxicity monitoring to detect harmful outputs or user crisis states.
Not certain from the listing — being a healthcare/therapy bot, it handles highly sensitive personal data but lacks explicit mention of HIPAA, GDPR, or robust identity/access management controls.
No multi-agent or marketplace interactions are described in the listing; it operates as a standalone conversational therapist.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).