Agno — agentic threat model
Agno is a powerful, open-source multi-agent framework whose primary security risks stem from the delegation of sandboxing, tool-execution safety, and state-management controls entirely to the implementing developer.
OWASP AIVSS score rationale
| Autonomy of Action | 0.70 | |
| Goal-Driven Planning | 0.80 | |
| Self-Modification | 0.30 | |
| Dynamic Tool Use | 0.80 | |
| Persistent Memory | 0.80 | |
| Contextual Awareness | 0.70 | |
| Dynamic Identity | 0.40 | |
| Multi-Agent Interactions | 0.90 | |
| Non-Determinism | 0.70 | |
| Opacity & Reflexivity | 0.60 |
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 — Agno supports various LLMs, making it susceptible to downstream model-level threats like adversarial prompt injection or misaligned outputs depending on the chosen foundation model.
Agno supports Agentic RAG and knowledge bases, exposing it to data poisoning, knowledge-base injection, and unauthorized data exfiltration via RAG pipelines.
As a Python-based agent framework, Agno directly handles memory management, tool integration, and multi-agent flows, making it highly vulnerable to tool misuse, memory poisoning, and insecure tool execution.
Not certain from the listing — Agno is an open-source Python framework, meaning deployment security (sandboxing, containerization, secret management for tools) is entirely up to the implementing developer.
Not certain from the listing — Agno does not explicitly detail built-in guardrails, evaluation suites, or observability logging in its high-level description, risking blind spots if not integrated externally.
Not certain from the listing — Being an open-source framework, compliance (e.g., SOC2, GDPR) and access control policies must be implemented at the application layer by the developer.
Agno supports multi-agent flow creation, introducing risks of cascading failures, trust abuse between agents, and complex coordination vulnerabilities within multi-agent setups.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).