awslabs.cloudwatch-mcp-server
Official AWS Labs CloudWatch MCP server. Runs locally (stdio) on the same host as the LLM client and uses the caller's AWS credentials to expose CloudWatch metrics, PromQL, alarms, and Logs Insights for AI-assisted troubleshooting. Announced July 2025 alongside the Application Signals MCP server.
One-click install for Cursor, VS Code, Claude, and 20+ other MCP clients, powered by API Commons MCP Install — visit install.apicommons.org for more information.
Documentation
Documentation link · transport stdio
Tools
get_metric_data— Retrieve detailed metric data with percentiles, math expressions, and multi-metric batching.get_metric_metadata— Obtain comprehensive metadata about a specific CloudWatch metric.get_recommended_metric_alarms— Suggest alarm configurations based on best practices and statistical analysis.analyze_metric— Examine metric data for trends, seasonality, and statistical properties.execute_promql_query— Run an instant PromQL query returning single-point-in-time values.execute_promql_range_query— Execute a PromQL query over a time window for trend analysis.get_promql_label_values— Retrieve values for a specific PromQL label for metric discovery.get_promql_series— Find time series matching PromQL label selectors.get_promql_labels— List available PromQL label names.get_active_alarms— Identify currently active alarms across the account.get_alarm_history— Retrieve historical state changes and patterns for alarms.describe_log_groups— Find metadata about log groups.analyze_log_group— Analyze logs for anomalies, patterns, and errors.execute_log_insights_query— Execute a Logs Insights query on log groups.execute_cwl_insights_batch— Run queries across multiple log groups/regions with automatic handling.get_logs_insight_query_results— Retrieve results from an executed Logs Insights query.cancel_logs_insight_query— Cancel an in-progress Logs Insights query.
About MCP
The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.
Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.