AI agents call aws_cloudwatch_metrics to retrieve information from Infra Ops without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves monitoring metrics from AWS CloudWatch, which is a read-only query operation. It queries existing metric data without creating, modifying, deleting, or executing any operations. The sibling tools demonstrate this server manages infrastructure resources, but this specific tool is limited to metric retrieval.
From the tool's definition Tool name 'aws_cloudwatch_metrics' and description 'Get CloudWatch metrics' indicate a retrieval operation with no data modification capability.
Attacks that exploit this kind of access
Get CloudWatch metrics. It is categorised as a Read tool in the Infra Ops MCP Server, which means it retrieves data without modifying state.
Register the Infra Ops MCP server in PolicyLayer and add a rule for aws_cloudwatch_metrics: 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 Infra Ops. Nothing to install.
aws_cloudwatch_metrics 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 aws_cloudwatch_metrics 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 aws_cloudwatch_metrics. 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.
aws_cloudwatch_metrics is provided by the Infra Ops MCP server (skyvanguard/infra-ops-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 →