AWS CLI MCP Server

What is AWS CLI MCP Server?

AWS CLI MCP Server is a Model Context Protocol (MCP) server that acts as a thin wrapper around the local AWS CLI installation, providing AI assistants with secure, controlled access to AWS services through the familiar AWS CLI interface. It is designed for users who want AI systems to manage AWS resources while leveraging existing AWS CLI configurations and credentials.

How to use AWS CLI MCP Server?

Install Python 3.10+ and the AWS CLI, then clone the repository, run pip install -e . to install dependencies, and start the server with python aws_mcp.py. The server runs in stdio mode and exposes a single tool, execute_aws_command_tool, which accepts an args array (e.g., ["sts", "get-caller-identity"]) and returns the command’s stdout, stderr, and exit code. Connect any MCP-compatible client (e.g., Claude Desktop) to the server via stdio transport.

Key features of AWS CLI MCP Server

• Thin wrapper that proxies commands directly to local AWS CLI
• Security by respecting existing AWS CLI credentials and permissions
• Compatibility with all AWS CLI commands and services
• Uses existing AWS CLI profiles and configurations
• Clean error handling and structured JSON output
• No additional authentication beyond existing AWS CLI setup

Use cases of AWS CLI MCP Server

• AI assistants managing AWS resources through natural language
• Automating routine AWS CLI tasks (e.g., listing instances, checking status)
• Enabling read-only AWS operations with guardrails from IAM policies
• Deploying or modifying infrastructure via the AWS CLI in a controlled manner
• Troubleshooting AWS services by executing CLI commands interactively

FAQ from AWS CLI MCP Server

What dependencies are required?

Python 3.10+ and a locally installed, configured AWS CLI with valid credentials (via aws configure, IAM roles, or environment variables).

What transport does the server use?

The server runs in stdio mode and communicates via the Model Context Protocol over standard input/output.

Does the server work with all AWS CLI commands?

Yes, it proxies any valid AWS CLI command and returns the output, so all services and commands supported by the local AWS CLI are accessible.

How are credentials and security handled?

The server uses the existing AWS CLI credentials and configuration—no additional authentication is needed. All AWS security mechanisms (IAM policies, MFA, etc.) remain in effect.

What is the name and format of the tool exposed?

The only tool is execute_aws_command_tool, which accepts a JSON array of arguments under the args key (e.g., {"args": ["s3", "ls"]}).