What is the AWS Knowledge Base Retrieval Server? It is an MCP server designed to fetch information from the AWS Knowledge Base via the Bedrock Agent Runtime.
How do you use it? First, set up your AWS credentials. Then, configure the server in your claude_desktop_config.json file. You can run it using Docker or an npx command, providing the necessary AWS access environment variables.
Its key feature is RAG (Retrieval-Augmented Generation). It retrieves context from the AWS Knowledge Base using a query and a specific Knowledge Base ID. A notable capability is support for fetching a customizable number of results.
Potential use cases include retrieving specific details from AWS documentation, integrating applications that need dynamic access to AWS Knowledge Base content, and enhancing customer support tools with quick access to AWS resources.
FAQ: How are AWS credentials set up? Obtain your AWS access key ID, secret access key, and region from the AWS Management Console, ensuring they have the required permissions. Can you customize the number of results? Yes, you can specify the number when making a query. Is there a license? Yes, it is licensed under the permissive MIT License.
{
"mcpServers": {
"aws-kb-retrieval": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-aws-kb-retrieval"
],
"env": {
"AWS_ACCESS_KEY_ID": "YOUR_ACCESS_KEY_HERE",
"AWS_SECRET_ACCESS_KEY": "YOUR_SECRET_ACCESS_KEY_HERE",
"AWS_REGION": "YOUR_AWS_REGION_HERE"
}
}
}
}