AgentReadyHomeAgent Listing

← Fast Wan

Fast Wan — agentic threat model

6.2AIVSS 6.2 · Medium

Fast Wan is a high-speed AI video generation platform with low agentic autonomy, posing primary risks around model abuse (e.g., deepfakes, policy bypass) and API resource exhaustion rather than autonomous decision-making vulnerabilities.

OWASP AIVSS score rationale

AIVSS = (CVSS_Base + AARS) × Mitigation_Factor, where AARS = (10 − CVSS_Base) × (Factor_Sum / 10) × ThM
CVSS base 5.3AARS uplift 0.94Factor sum 2.0/10Threat ×1.0Mitigation ×1.0
Autonomy of Action
0.10
Goal-Driven Planning
0.00
Self-Modification
0.00
Dynamic Tool Use
0.10
Persistent Memory
0.10
Contextual Awareness
0.20
Dynamic Identity
0.00
Multi-Agent Interactions
0.00
Non-Determinism
0.80
Opacity & Reflexivity
0.70

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.

L1 · Foundation Models✓ mapped

Leverages FastWan 2.2 and 2.1 sparse distillation models. Primary threats include adversarial prompt injection to bypass safety filters, model extraction/stealing via API harvesting, and the generation of harmful, copyrighted, or deepfake content.

L2 · Data Operations⚠ not certain from listing

Not certain from the listing — The dataset used for training or fine-tuning the Wan models is not disclosed. Potential threats include training data copyright infringement, lack of data lineage/provenance, and potential data poisoning if user-uploaded images/videos are used for fine-tuning.

L3 · Agent Frameworks⚠ not certain from listing

Not certain from the listing — The platform appears to function as a direct inference API rather than a complex agentic framework. Orchestration threats are minimal, though insecure API integration in downstream client applications could lead to prompt injection vulnerabilities.

L4 · Deployment & Infrastructure⚠ not certain from listing

Not certain from the listing — The GPU hosting environment and API sandboxing mechanisms are not detailed. Key threats include GPU resource exhaustion (DoS) due to the high-speed generation capabilities, and API key exposure/theft leading to unauthorized billing.

L5 · Evaluation & Observability⚠ not certain from listing

Not certain from the listing — There is no mention of output moderation guardrails, input filtering, or generation logging. The lack of observability tools increases the risk of undetected generation of policy-violating or malicious video content.

L6 · Security & Compliance (cross-cutting)⚠ not certain from listing

Not certain from the listing — No details are provided regarding user authentication, access controls, or compliance with regulations like the EU AI Act (specifically regarding watermarking and labeling of AI-generated synthetic media).

L7 · Agent Ecosystem⚠ not certain from listing

Not certain from the listing — The platform is offered as an API and horizontal tool, but direct multi-agent or marketplace interactions are not described. Downstream integration risks include cascading failures in client applications if the Fast Wan API experiences outages.

MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).