AI agents call vunet_query_metric to retrieve information from Vunet without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves observability data (metrics) from a multi-tenant platform without side effects. Querying metrics is a read operation with minimal blast radius—the worst misuse would be data exfiltration, but not data modification or deletion. The observability context (metrics, traces, logs from Vunet) confirms this is a passive data retrieval interface.
From the tool's definition Tool name 'vunet_query_metric' and description 'Query a Vunet data model/metric' indicate a read-only operation. The description explicitly states it queries metrics with time ranges and filters, with no mention of modifying, deleting, or executing operations.
Attacks that exploit this kind of access
Query a Vunet data model/metric with flexible time ranges and filters. Automatically uses the configured tenant. It is categorised as a Read tool in the Vunet MCP Server, which means it retrieves data without modifying state.
Register the Vunet MCP server in PolicyLayer and add a rule for vunet_query_metric: 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 Vunet. Nothing to install.
vunet_query_metric 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 vunet_query_metric 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 vunet_query_metric. 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.
vunet_query_metric is provided by the Vunet MCP server (mithung-vunet/vunet-mcp-server). 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 →