Tome Health — agentic threat model
Tome Health presents a high-consequence privacy and safety risk due to its processing of sensitive genetic and blood test data, where LLM hallucinations could translate to incorrect medical interpretations or severe PHI leaks.
OWASP AIVSS score rationale
| Autonomy of Action | 0.20 | |
| Goal-Driven Planning | 0.30 | |
| Self-Modification | 0.00 | |
| Dynamic Tool Use | 0.10 | |
| Persistent Memory | 0.40 | |
| Contextual Awareness | 0.60 | |
| Dynamic Identity | 0.00 | |
| Multi-Agent Interactions | 0.00 | |
| Non-Determinism | 0.50 | |
| Opacity & Reflexivity | 0.40 |
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 commercial LLMs to translate medical jargon into gamified text. Risks include prompt injection leading to incorrect medical advice or hallucinations about genetic risks.
Not certain from the listing — ingests highly sensitive PHI (blood tests, genetic data). Risks include data exfiltration, unauthorized access to genetic profiles, and lack of secure RAG/data isolation.
Not certain from the listing — orchestrates gamified 'quests' and 'badges' based on health data. Risks include insecure state management of user progress and manipulation of the gamification logic.
Not certain from the listing — requires highly secure hosting (e.g., HIPAA-compliant cloud). Risks include insecure storage of raw genetic files and lack of strict access controls.
Not certain from the listing — critical need for clinical validation guardrails to prevent harmful medical hallucinations. Risks include lack of automated drift detection for medical terminology.
Not certain from the listing — must comply with HIPAA, GDPR, and health data regulations. Risks include non-compliance, lack of robust audit logs for PHI access, and weak user authentication.
The listing describes a standalone horizontal health application with no multi-agent or marketplace integrations mentioned; ecosystem risks are currently negligible.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology. Are you the vendor? Factual corrections are free.