11Sight — agentic threat model
11Sight presents a moderate-to-high risk profile due to its direct integration with customer booking systems, CRM databases, and outbound telephony channels. A compromise could lead to widespread PII exposure, unauthorized appointment manipulation, and reputational damage via automated spam calls.
OWASP AIVSS score rationale
| Autonomy of Action | 0.70 | |
| Goal-Driven Planning | 0.50 | |
| Self-Modification | 0.10 | |
| Dynamic Tool Use | 0.60 | |
| Persistent Memory | 0.50 | |
| Contextual Awareness | 0.50 | |
| Dynamic Identity | 0.20 | |
| Multi-Agent Interactions | 0.30 | |
| 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 — the underlying LLM or foundation model is not specified. Threats include prompt injection leading to unauthorized actions (e.g., booking fake appointments) or jailbreaks bypassing dealership policies.
Not certain from the listing — details about the vector database or RAG architecture for dealership hours, inventory, and recall data are omitted. Threats include knowledge-base poisoning (e.g., altering dealership hours or recall details) and PII leakage from customer interaction history.
The agent orchestrates conversational flows, appointment booking APIs, and outbound calling. Threats include insecure tool integration (e.g., API abuse via booking tools) and state manipulation during rescheduling or cancellation flows.
Not certain from the listing — hosting environment (SaaS, cloud provider) is not detailed. Threats include unauthorized access to API keys for telephony/CRM integrations and container escape.
Not certain from the listing — monitoring and guardrail frameworks are not specified. Threats include undetected drift in conversational quality or failure to log malicious prompt injection attempts.
Not certain from the listing — compliance standards (e.g., SOC2, GDPR for customer PII) are not mentioned. Threats include lack of robust RBAC for dealership staff accessing the agent's backend and insufficient audit trails for automated bookings.
The agent suite consists of multiple specialized roles (Front-desk, Service, Sales, Outbound) and integrates with external CRMs and telephony. Threats include cascading failures if the outbound agent is compromised to spam customers, or unauthorized data sharing between the sales and service agents.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).