AgentReadyHomeAgent Listing

← OpenRouter AI

OpenRouter AI — agentic threat model

8.9AIVSS 8.9 · High

OpenRouter AI acts as a centralized API aggregator and router for multiple LLMs, presenting a high-value target for API key theft and data interception, though its direct agentic autonomy and planning risks are low.

OWASP AIVSS score rationale

AIVSS = (CVSS_Base + AARS) × Mitigation_Factor, where AARS = (10 − CVSS_Base) × (Factor_Sum / 10) × ThM
CVSS base 8.5AARS uplift 0.38Factor sum 2.5/10Threat ×1.0Mitigation ×1.0
Autonomy of Action
0.10
Goal-Driven Planning
0.10
Self-Modification
0.00
Dynamic Tool Use
0.20
Persistent Memory
0.20
Contextual Awareness
0.40
Dynamic Identity
0.20
Multi-Agent Interactions
0.10
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.

L1 · Foundation Models✓ mapped

OpenRouter aggregates multiple third-party LLMs. Threats include model-specific vulnerabilities, adversarial prompt injection, and misaligned outputs from the underlying models that OpenRouter routes to.

L2 · Data Operations⚠ not certain from listing

Not certain from the listing — OpenRouter acts as a proxy, but data logging, caching, or RAG capabilities are not detailed. Potential threats include data exfiltration of prompt/response payloads or insecure caching of sensitive images/charts.

L3 · Agent Frameworks⚠ not certain from listing

Not certain from the listing — OpenRouter is primarily an API router rather than an orchestration framework, but any internal routing logic or fallback mechanisms could suffer from framework-level manipulation or routing bypasses.

L4 · Deployment & Infrastructure⚠ not certain from listing

Not certain from the listing — The infrastructure hosting the API gateway and model routing must secure API keys and user data in transit. Threats include API key leakage, man-in-the-middle attacks, or container compromise of the routing gateway.

L5 · Evaluation & Observability⚠ not certain from listing

Not certain from the listing — It is unclear what guardrails, logging, or anomaly detection OpenRouter implements to monitor malicious payloads or abuse of the aggregated models.

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

Not certain from the listing — Compliance controls, data privacy policies (e.g., GDPR/CCPA regarding prompt data retention), and IAM/access controls for API keys are not specified in the directory listing.

L7 · Agent Ecosystem✓ mapped

OpenRouter acts as a central hub/marketplace for LLMs. Threats include cascading failures if a major upstream model provider goes down, or malicious models being integrated into the router.

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