AI agents use compare_models to commit financial operations through OpenRouter MCP Server — usually the final step of a payment, billing, or trading workflow. A call moves real money.
Invoking multiple models in parallel multiplies API usage charges against the user's OpenRouter account. While the tool's primary action is Execute (running inference across models), the financial side-effect of accumulating costs across several models on every call elevates this to Financial severity. A misused or looping agent could rapidly exhaust credits.
From the tool's definition Compare responses from multiple models — calls multiple AI models simultaneously via OpenRouter, each of which incurs per-token API costs billed to the user's account.
Documented attack patterns abuse exactly the kind of access compare_models gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and OpenRouter MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for compare_models:
{
"version": "1",
"default": "deny",
"tools": {
"compare_models": {
"deny_if": [
{
"conditions": [],
"on_deny": "Requires human approval."
}
]
}
}
}Any call to compare_models is blocked until a human approves it. The rest of the server keeps working.
Free to start. No card required.
Compare responses from multiple models. It is categorised as a Financial tool in the OpenRouter MCP Server MCP Server, which means it involves financial transactions. Block by default and require explicit approval.
Register the OpenRouter MCP Server MCP server in PolicyLayer and add a rule for compare_models: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches OpenRouter MCP Server. Nothing to install.
compare_models is a Financial tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the compare_models rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for compare_models. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
compare_models is provided by the OpenRouter MCP Server MCP server (th3nolo/openrouter-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from OpenRouter MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
4 OpenRouter MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.