Mendable — agentic threat model
Mendable presents a moderate-to-high risk profile primarily centered on data privacy and integrity due to its data ingestion and continuous training features. While it claims enterprise-grade security, the centralized hosting of custom chat applications makes it an attractive target for data exfiltration and model poisoning attacks.
OWASP AIVSS score rationale
| Autonomy of Action | 0.30 | |
| Goal-Driven Planning | 0.20 | |
| Self-Modification | 0.10 | |
| Dynamic Tool Use | 0.30 | |
| Persistent Memory | 0.40 | |
| Contextual Awareness | 0.60 | |
| Dynamic Identity | 0.10 | |
| Multi-Agent Interactions | 0.20 | |
| 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.
Supports multiple foundation models and customization. Risks include adversarial prompt injection bypassing chatbot guardrails, and potential model stealing or membership inference if custom-trained weights are exposed.
Features data ingestion and continuous training. This introduces significant risks of training data poisoning, knowledge-base contamination, and unauthorized data exfiltration of ingested enterprise documents.
Not certain from the listing — the orchestration framework for the chat applications is not detailed, but insecure tool integration or prompt leakage could occur if the framework lacks strict input/output sanitization.
Not certain from the listing — details on the hosting environment for the 'one-line deployment' are omitted, raising potential concerns regarding container isolation, tenant sandboxing, and secure API credential storage.
Not certain from the listing — while continuous training is supported, it is unclear what observability, drift detection, or automated evaluation guardrails are in place to monitor chatbot outputs.
Explicitly emphasizes 'enterprise-grade security'. However, specific compliance certifications (e.g., SOC 2, ISO 27001) or granular access control mechanisms are not detailed in the public listing.
Not certain from the listing — there is no mention of multi-agent orchestration, marketplace integrations, or agent-to-agent communication protocols.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).