75way Technologies — agentic threat model
As a custom development provider, 75way Technologies builds bespoke AI agents with deep API and enterprise integrations, presenting a variable risk profile that depends heavily on client-specific implementations and the rigor of their custom security controls.
OWASP AIVSS score rationale
| Autonomy of Action | 0.50 | |
| Goal-Driven Planning | 0.40 | |
| Self-Modification | 0.30 | |
| Dynamic Tool Use | 0.50 | |
| Persistent Memory | 0.40 | |
| Contextual Awareness | 0.50 | |
| Dynamic Identity | 0.20 | |
| Multi-Agent Interactions | 0.30 | |
| Non-Determinism | 0.40 | |
| 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 listing describes custom AI agent development services but does not specify which foundation models (e.g., GPT-4, Claude, Llama) are used. Threats include model-specific vulnerabilities like adversarial prompt injection or data poisoning depending on the client's choice of LLM.
Not certain from the listing — While they mention 'Data security & compliance' and 'data analytics', the specific vector databases, RAG pipelines, or training data operations are custom-built per client. Threats include knowledge-base poisoning or data exfiltration if client integrations lack strict access controls.
Not certain from the listing — They offer 'Workflow automation & process optimization' and 'Adaptive learning & behavior optimization', indicating orchestration capabilities, but the specific framework (LangChain, AutoGen, CrewAI, or proprietary) is not disclosed. Threats include insecure tool integration and tool misuse in custom-built workflows.
Not certain from the listing — They provide 'deployment, and support' for mobile, IoT, and blockchain, but the hosting infrastructure (AWS, Azure, on-premise) and sandboxing mechanisms for executed code are determined on a project-by-project basis. Threats include container compromise or privilege escalation in custom deployments.
Not certain from the listing — The listing does not detail specific evaluation frameworks, guardrails, or continuous monitoring tools used for the deployed agents. Gaps here could lead to undetected drift or prompt injection attacks.
Not certain from the listing — They claim 'Data security & compliance' as a key feature, but do not specify compliance standards (like SOC2, ISO 27001, or HIPAA) or specific identity/authorization controls implemented across their custom solutions.
Not certain from the listing — They build custom agents and blockchain platforms, which could involve multi-agent coordination or marketplace interactions, but no specific ecosystem architecture is detailed. Threats include cascading failures or trust abuse in multi-agent workflows.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).