AI agents call evaluate_response_quality to retrieve information from Msty Admin MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs evaluation and analysis of responses, which is a read-only operation. It retrieves or assesses data without side effects, modifications, or destructive actions. The low severity reflects that misuse would only result in incorrect assessments, not data loss or system compromise.
From the tool's definition Tool name is 'evaluate_response_quality' and the description states it will 'Evaluate response quality.' This is an assessment/analysis operation with no ability to modify, delete, or execute external code.
Documented attack patterns abuse exactly the kind of access evaluate_response_quality gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Msty Admin MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for evaluate_response_quality:
{
"version": "1",
"default": "deny",
"tools": {
"evaluate_response_quality": {}
}
}evaluate_response_quality is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Evaluate response quality. It is categorised as a Read tool in the Msty Admin MCP MCP Server, which means it retrieves data without modifying state.
Register the Msty Admin MCP server in PolicyLayer and add a rule for evaluate_response_quality: 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 Msty Admin MCP. Nothing to install.
evaluate_response_quality 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 evaluate_response_quality 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 evaluate_response_quality. 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.
evaluate_response_quality is provided by the Msty Admin MCP server (m-pineapple/msty-admin-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Msty Admin MCP, 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.
36 Msty Admin MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.