CloudWatch

Provides CloudWatch log groups, alarms, and Logs Insights queries via MCP for easy access

python

4

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "charliefng-cloudwatch-mcp": {
      "command": "python",
      "args": [
        "cloudwatch_server.py"
      ],
      "env": {
        "AWS_REGION": "us-west-2",
        "AWS_ACCESS_KEY_ID": "YOUR_ACCESS_KEY",
        "AWS_SECRET_ACCESS_KEY": "YOUR_SECRET_KEY"
      }
    }
  }
}

You can use the CloudWatch MCP Server to interact with AWS CloudWatch resources through the MCP protocol. It exposes log groups, alarms, and Logs Insights queries as accessible resources and tools, letting you list, inspect, and query CloudWatch data from a local MCP client or from an MCP-based workflow.

How to use

To use this MCP server, run it from your environment and connect with an MCP client. You can inspect resources such as log groups and alarms, and you can run queries across multiple log groups using CloudWatch Logs Insights. Use the provided tools to discover available fields, check whether specific log groups exist, and fetch saved queries. The server is designed to parse JSON messages in log data automatically and to present structured results that are easy to consume in MCP-enabled tooling.

How to install

Prerequisites and setup are as follows. Install Python 3.12 or higher, ensure AWS credentials are configured, and have the MCP CLI available (version 0.1.1 or higher). Then you can run the server directly or via the MCP CLI.

Step-by-step commands you will use:

Additional configuration and notes

Run the server directly with Python or via the MCP CLI. The server exposes resources for CloudWatch log groups, alarms, and saved queries, and provides tools to query logs, discover fields, check log group existence, and fetch saved queries. No additional configuration is required beyond your AWS credentials and network access to AWS services.

Examples and practical usage

Start the MCP server using Python: python cloudwatch_server.py.

Start the MCP server using the MCP CLI: mcp run cloudwatch_server.py.

Troubleshooting and tips

Ensure AWS credentials are accessible to the environment running the server. If you encounter permission errors, verify that your IAM user or role has the necessary CloudWatch permissions. If the server cannot reach AWS endpoints, confirm network access and region configuration.

Tools and endpoints you can use with the server

query_logs — Query CloudWatch logs using CloudWatch Insights across one or more log groups. It automatically parses JSON in the @message field and returns structured data.

Supported resources and paths

cloudwatch://log-groups — Lists all CloudWatch log groups.

Testing the CloudWatch integration

You can test by listing resources or running queries through the MCP client. Use the provided test utilities to verify log groups, alarms, and saved queries.

Available tools

query_logs

Query CloudWatch logs using CloudWatch Insights across one or more log groups. Automatically parses JSON in the @message field and returns structured results.

discover_log_fields

Discover available fields across multiple log groups, assuming shared schema. Detects nested JSON fields in @message and identifies field types.

log_group_exists

Check whether specified CloudWatch log groups exist and return a map of existence per log group.

get_saved_queries

Fetch all saved CloudWatch Logs Insights queries with no parameters.