mcp-registry/awslabs/mcp

    ==================
      
       /// MCP ///
      /// AWS ///
        
    ==================
        
    [server:online]
    [protocol:ready]

awslabs/mcp

by awslabs

Suite of AWS-focused Model Context Protocol (MCP) servers – lightweight Python services that expose AWS tooling, documentation, pricing, IaC and other capabilities to MCP-compatible AI clients (Amazon Q, Cline, Cursor, Windsurf, etc.).

4.7k
585
Open Source
GitHubWebsiteDocumentation

Installation

1. Prerequisites
• Python ≥3.9
• AWS account with credentials configured in ~/.aws/credentials or via environment variables (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_DEFAULT_REGION)
• (Optional) Docker if you prefer container deployment
2. Clone the repository
git clone https://github.com/awslabs/mcp.git cd mcp
3. Create virtual environment & install dependencies
python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate pip install --upgrade pip pip install -r requirements.txt
(If a `requirements-dev.txt` file exists, install it the same way for local testing.)
4. Build & run the MCP server locally
# Update the config file (see ./config/example-config.yaml) with your AWS settings cp config/example-config.yaml config/config.yaml vi config/config.yaml # or your preferred editor # Start the server python -m mcp.server --config ./config/config.yaml
5. Using Docker (alternative)
docker build -t aws-mcp:latest . docker run -p 8080:8080 -e AWS_ACCESS_KEY_ID -e AWS_SECRET_ACCESS_KEY aws-mcp:latest
6. Verify
Open http://localhost:8080/health or run:
curl http://localhost:8080/health
7. Install CLI / client library (optional)
pip install aws-mcp-client # if published to PyPI
8. Deployment to AWS (optional sketch)
• Review the CDK/Terraform templates in `deploy/`
• Configure parameters (VPC, sub-nets, domain, certificate)
• Deploy: `cd deploy/cdk && cdk deploy` or follow README in the deploy folder.

Documentation

License: Apache License 2.0
Updated 7/15/2025