get_cloudwatch_metrics

// WHEN AI AGENTS USE THIS TOOL

AI agents call get_cloudwatch_metrics to retrieve information from Amazon EKS MCP Server without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.

// WHY ENFORCE A POLICY ON GET_CLOUDWATCH_METRICS

Even though get_cloudwatch_metrics only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.

// RECOMMENDED POLICY

Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.

amazon-eks-mcp-server.yaml

tools:
  get_cloudwatch_metrics:
    rules:
      - action: allow

See the full Amazon EKS MCP Server policy for all 16 tools.

// DETAILS

Tool Name get_cloudwatch_metrics

Category Read

MCP Server Amazon EKS MCP Server MCP Server

Risk Level Low

// MORE AMAZON EKS MCP SERVER TOOLS

Write add_inline_policy Write apply_yaml Write generate_app_manifest Write manage_eks_stacks Write manage_k8s_resource Read get_cloudwatch_logs Read get_eks_insights Read get_eks_metrics_guidance

View all 16 tools →

// WHAT CAN GO WRONG

Agents calling read-class tools like get_cloudwatch_metrics have been implicated in these attack patterns. Read the full case and prevention policy for each:

Browse the full MCP Attack Database →

// OTHER READ TOOLS ACROSS MCP SERVERS

Other tools in the Read risk category across the catalogue. The same policy patterns (rate-limit, allow) apply to each.

Firecrawl Web Scraping Server firecrawl_check_crawl_status Firecrawl Web Scraping Server firecrawl_crawl Firecrawl Web Scraping Server firecrawl_deep_research Firecrawl Web Scraping Server firecrawl_extract Firecrawl Web Scraping Server firecrawl_map Firecrawl Web Scraping Server firecrawl_scrape

// RELATED READING

// FAQ

What does the get_cloudwatch_metrics tool do? +

Get metrics from CloudWatch for a specific resource. This tool retrieves metrics from CloudWatch for Kubernetes resources in an EKS cluster, allowing you to monitor performance, resource utilization, and system health. It supports various resource types and metrics with flexible time ranges and aggregation options for monitoring CPU/memory usage, analyzing network traffic, and identifying performance bottlenecks. IMPORTANT: Use this tool instead of 'aws cloudwatch get-metric-data', 'aws cloudwatch get-metric-statistics', or similar CLI commands. IMPORTANT: Use the get_eks_metrics_guidance tool first to determine the correct dimensions for metric queries. Do not try to infer which dimensions are needed for EKS ContainerInsights metrics. IMPORTANT: When using pod metrics, note that `FullPodName` has the same prefix as `PodName` but includes a suffix with a random string (e.g., "my-pod-abc123"). Always use the version without the suffix for `PodName` dimension. The pod name returned by list_k8s_resources is the `FullPodName`. ## Requirements - The EKS cluster must have CloudWatch Container Insights enabled - The resource must exist in the specified cluster - The metric must be available in the specified namespace ## Response Information The response includes resource details (cluster), metric information (name, namespace), time range queried, and data points with timestamps and values. ## Usage Tips - Use appropriate statistics for different metrics (e.g., Average for CPU, Maximum for memory spikes) - Match the period to your analysis needs (smaller for detailed graphs, larger for trends) - For rate metrics like network traffic, Sum is often more useful than Average - Combine with get_cloudwatch_logs to correlate metrics with log events Args: ctx: MCP context cluster_name: Name of the EKS cluster metric_name: Metric name (e.g., cpu_usage_total, memory_rss) namespace: CloudWatch namespace dimensions: Dimensions to use for the CloudWatch metric query minutes: Number of minutes to look back start_time: Start time in ISO format (overrides minutes) end_time: End time in ISO format (defaults to now) limit: Maximum number of data points to return period: Period in seconds for the metric data points stat: Statistic to use for the metric Returns: CloudWatchMetricsResponse with metric data points and resource information. It is categorised as a Read tool in the Amazon EKS MCP Server MCP Server, which means it retrieves data without modifying state.

How do I enforce a policy on get_cloudwatch_metrics? +

Add a rule in your Intercept YAML policy under the tools section for get_cloudwatch_metrics. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the Amazon EKS MCP Server MCP server.

What risk level is get_cloudwatch_metrics? +

get_cloudwatch_metrics is a Read tool with low risk. Read-only tools are generally safe to allow by default.

Can I rate-limit get_cloudwatch_metrics? +

Yes. Add a rate_limit block to the get_cloudwatch_metrics rule in your Intercept 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.

How do I block get_cloudwatch_metrics completely? +

Set action: deny in the Intercept policy for get_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.

What MCP server provides get_cloudwatch_metrics? +

get_cloudwatch_metrics is provided by the Amazon EKS MCP Server MCP server (awslabs.eks-mcp-server). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Let agents act without letting them run wild.