Get the single best-fit LLM for a task, with pricing and reasoning.
AI agents call recommend_model to retrieve information from Whichmodel without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries cost and capability data to provide recommendations. It does not create, modify, execute code, delete data, or commit financial transactions. It is purely informational, analogous to a search or lookup function. Even in the context of a model routing advisor, the tool itself performs no execution, no data modification, and no financial commitment—it only retrieves and calculates recommendations.
From the tool's definition Tool 'recommend_model' returns advisory data (pricing and reasoning) based on task input. The description states 'Get the single best-fit LLM' — a retrieval and analysis operation with no side effects.
Attacks that exploit this kind of access
Get the single best-fit LLM for a task, with pricing and reasoning. It is categorised as a Read tool in the Whichmodel MCP Server, which means it retrieves data without modifying state.
Register the Whichmodel MCP server in PolicyLayer and add a rule for recommend_model: 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 Whichmodel. Nothing to install.
recommend_model is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the recommend_model 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 recommend_model. 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.
recommend_model is provided by the Whichmodel MCP server (simonamsharp/routewise-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →