AWS Cloud MCP Server
Bridges AI assistants directly to the AWS CLI with pipeline execution, jq/grep filtering, and automatic --help documentation discovery.
About this Server
AWS Cloud MCP Server (`aws-mcp` / `cloud-mcp-server`) empowers AI assistants like Claude Code, Cursor, and VS Code to interact with all Amazon Web Services through the AWS CLI. Rather than wrapping separate APIs, it provides two high-leverage primitives: on-demand command documentation lookup via `aws_cli_help` and chained command execution with shell pipelines (`jq`, `grep`) via `aws_cli_pipeline`, bound to IAM policies.
Direct low-latency RPC protocol communication with structured JSON schemas.
Restricted capability bounds to prevent accidental modifications or leakage.
Security & Best Practices
When to Use (and When Not To)
- •Querying EC2, S3, RDS, ECS, Lambda, or CloudWatch metrics directly from AI coding assistants
- •Running complex AWS command pipelines with jq parsing to inspect cloud infrastructure state
- •Learning unknown AWS CLI commands on-demand without manual browser documentation searches
- •Kubernetes kubectl cluster and Helm deployments (use Kubernetes MCP Server instead)
- •Ephemeral disposable cloud browsers and verification inboxes (use Ausca MCP instead)
- •External SaaS status page health verification (use API Status Check MCP instead)
Install & Setup
claude mcp add aws -- uvx aws-mcpExposed Tools & Endpoints
2 toolsFunctions your AI agent can invoke via this Model Context Protocol server
Executes AWS CLI commands with optional shell pipeline filters (jq, grep, awk) to query and manage AWS infrastructure.
Retrieves comprehensive documentation, parameters, and syntax examples for any AWS CLI command.