TL;DR: Model routing and agent policy address different controls, Archestra says, with routing selecting the cheapest capable model and policy governing tool access, visible context, and turn audit. Agent turns need separate evidence for routing and enforcement if audit trails are to stay trustworthy.
Editorial analysis by NHI Mgmt Group, based on content published by Archestra: “Martian router Archestra: cost-aware model routing, identity-aware tool policy”.
Key questions
Q: Why does cheaper model selection not solve agent security risk?
A: Cheaper model selection only affects inference spend and model fit.
Q: What breaks when a router is treated as the only gateway for tools?
A: The tool surface becomes effectively ungoverned because the routing layer lacks per-user entitlement, sandboxing, and turn-level policy enforcement.
Q: How should teams separate model routing from agent access control?
A: Treat model routing as a cost and performance function, and treat agent access control as an identity and policy function.
Practitioner guidance
- Separate routing from authorisation Keep model-selection logic in one service and per-identity tool policy in another so cost optimisation cannot expand access scope.
- Enforce turn-level tool allowlists Resolve the tools visible to the agent from the caller's identity and session context before any invocation is exposed.
- Quarantine untrusted tool output Route external or adversarial tool results through a containment step before they can modify the agent's active context.
Bottom line: Model routing and agent governance solve different problems, and mixing them creates a false sense of control.
Glossary terms: Model Router, Agent Turn, MCP Gateway, Prompt Injection (Agentic)
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
Cost-aware routing and agent governance are orthogonal control problems. One optimises inference spend, the other constrains what an identity-bound agent can see and do. The article correctly separates those responsibilities, and that separation is the point: a cheaper model does not reduce the need for per-turn authorisation, MCP scoping, or auditable tool access. Practitioners should read this as a warning against collapsing performance controls into identity controls.
A question worth separating out:
Q: What should security teams audit in an agent stack with a model router?
A: Audit the identity of the caller, the tool list exposed on that turn, the tools actually invoked, and the model chosen for the request. Those records show whether a failure happened in routing, authorisation, or context handling, which are separate control problems.
👉 Read our full editorial: Model routing and agent policy are separate control layers