AgentReadyHomeAgent Listing

← Soulmate IO

Soulmate IO — agentic threat model

6.6AIVSS 6.6 · Medium

Soulmate IO presents low operational agentic risk due to its lack of external tool execution or autonomous planning, but carries high privacy risks due to its reliance on persistent memory storing sensitive, personal user conversations.

OWASP AIVSS score rationale

AIVSS = (CVSS_Base + AARS) × Mitigation_Factor, where AARS = (10 − CVSS_Base) × (Factor_Sum / 10) × ThM
CVSS base 5.3AARS uplift 1.69Factor sum 3.6/10Threat ×1.0Mitigation ×0.95
Autonomy of Action
0.20
Goal-Driven Planning
0.10
Self-Modification
0.20
Dynamic Tool Use
0.10
Persistent Memory
0.80
Contextual Awareness
0.60
Dynamic Identity
0.10
Multi-Agent Interactions
0.00
Non-Determinism
0.70
Opacity & Reflexivity
0.80

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 — The underlying LLMs and text-to-speech (TTS) models are proprietary and closed-source, making them vulnerable to black-box adversarial prompt injection or model evasion attacks that could alter the companion's behavior.

L2 · Data Operations⚠ not certain from listing

Not certain from the listing — The platform relies on persistent memory to store user preferences and chat history. Without explicit details on encryption or vector database security, this data is vulnerable to unauthorized access, data exfiltration, or memory poisoning.

L3 · Agent Frameworks⚠ not certain from listing

Not certain from the listing — The orchestration framework managing the dialogue state and memory retrieval is unspecified, presenting potential risks of state manipulation or context-stuffing vulnerabilities.

L4 · Deployment & Infrastructure⚠ not certain from listing

Not certain from the listing — The hosting infrastructure for the web/mobile application and voice synthesis APIs is not detailed, leaving standard cloud infrastructure threats (e.g., API exposure, insecure endpoints) unassessed.

L5 · Evaluation & Observability⚠ not certain from listing

Not certain from the listing — There is no mention of real-time guardrails, output filtering for sensitive/harmful content, or conversational drift monitoring to prevent toxic or manipulative AI behavior.

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

Not certain from the listing — While marketed as a 'private and secure' platform, no specific compliance standards (such as GDPR, SOC2) or data deletion mechanisms are detailed to verify these privacy claims.

L7 · Agent Ecosystem✓ mapped

The agent operates as a closed, isolated 1-on-1 companion platform with no multi-agent collaboration, marketplace integrations, or external agent-to-agent communication channels described.

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