OutSystems — agentic threat model
OutSystems AI Agent Builder provides an enterprise-grade low-code platform for deploying RAG-powered agents, presenting a moderate-to-high risk profile due to deep integration with full-stack enterprise applications, balanced by robust built-in governance and observability controls.
OWASP AIVSS score rationale
| Autonomy of Action | 0.50 | |
| Goal-Driven Planning | 0.60 | |
| Self-Modification | 0.20 | |
| Dynamic Tool Use | 0.70 | |
| Persistent Memory | 0.50 | |
| Contextual Awareness | 0.70 | |
| Dynamic Identity | 0.40 | |
| Multi-Agent Interactions | 0.30 | |
| Non-Determinism | 0.70 | |
| 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.
The platform is model-provider agnostic, meaning it inherits the specific vulnerabilities of whichever third-party LLM is integrated, including prompt injection, model-side data leakage, and service availability risks.
Features custom agent development powered by RAG. This introduces risks of knowledge-base poisoning, unauthorized data retrieval through semantic search, and data exfiltration if access controls on the underlying vector stores are misconfigured.
Agents are built using OutSystems' low-code orchestration. Vulnerabilities include insecure tool integration, logic flaws in the low-code visual workflows, and prompt injection attacks that manipulate the agent's decision-making paths.
Not certain from the listing — OutSystems typically deploys on its own cloud infrastructure or on-premises. Infrastructure risks include container isolation failures, insecure API gateways, and inadequate sandboxing of custom code execution environments.
Includes built-in 'Agent Monitoring and Observability' features, which help mitigate risks by tracking agent execution, logging LLM interactions, and detecting anomalous behaviors or drift.
Emphasizes IT governance, standardization, and security controls to prevent shadow AI, ensuring that deployed agents align with enterprise compliance frameworks and access control policies.
Provides a library of quick-start generative AI apps. This introduces ecosystem risks such as template supply-chain vulnerabilities, insecure default configurations in pre-built apps, and potential cascading failures if multiple quick-start apps are chained together.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).