AI agents use add_metric to create or update resources in Querywise — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Querywise environment.
The tool creates or adds a metric to the semantic layer metadata store, modifying the data dictionary/glossary structure. This is a Write operation (reversible via delete_metric) rather than Read. Severity is medium because misconfigured metrics could mislead query results or analytics, but impact is scoped to metadata rather than production data.
From the tool's definition Tool name 'add_metric' indicates creation/modification of metadata; server description mentions 'metrics' as part of semantic layer; sibling tools include 'delete_metric' confirming metrics are persistent data structures; description is empty, limiting…
Attacks that exploit this kind of access
add_metric. It is categorised as a Write tool in the Querywise MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Querywise MCP server in PolicyLayer and add a rule for add_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 Querywise. Nothing to install.
add_metric is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the add_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 add_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.
add_metric is provided by the Querywise MCP server (kosminus/querywise-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 →