Retrieve AWS Knowledge Base Data
AWS KB Retrieval MCP Server queries an AWS Bedrock Knowledge Base by search query and Knowledge Base ID, returning RAG context results.
1.0.0Add to Favorites
Why it matters
Leverage AWS Knowledge Base for RAG to retrieve relevant information. This asset enables context retrieval from your AWS KB based on user queries.
Outcomes
What it gets done
Connect to AWS Knowledge Base
Retrieve context using RAG
Customize number of results
Search AWS KB with provided query
Source
Get it from source
Spark does not host a copy of it.
Open sourceReports
Agent outcome reports
No reports yet
Capabilities
Tools your agent gets
Perform retrieval operations using the AWS Knowledge Base with customizable result count.
Overview
AWS KB Retrieval MCP Server
A single-tool MCP server that retrieves RAG context from an AWS Bedrock Knowledge Base via the Bedrock Agent Runtime, given a query and a Knowledge Base ID. Reach for it when an MCP client needs to query an existing populated AWS Bedrock Knowledge Base for grounding context; note it ships from Anthropic's archived reference-servers repository, not an actively maintained standalone project.
What it does
AWS KB Retrieval MCP Server is a single-purpose MCP server that retrieves context from an AWS Bedrock Knowledge Base via the Bedrock Agent Runtime, for retrieval-augmented generation (RAG). It exposes one tool, retrieve_from_aws_kb, that runs a retrieval query against a specified Knowledge Base ID and returns a configurable number of results.
When to use - and when NOT to
Use it when you want an MCP client to pull grounding context from an existing AWS Bedrock Knowledge Base by query, instead of hand-implementing Bedrock Agent Runtime retrieval calls yourself. This connector lives in the official modelcontextprotocol/servers-archived repository - it has been archived upstream, so treat it as a reference implementation rather than an actively maintained project. It assumes you already have a populated Bedrock Knowledge Base; it doesn't create, ingest into, or manage one, and it requires AWS credentials (access key, secret key, region) with Bedrock Agent Runtime permissions.
Capabilities
retrieve_from_aws_kb takes a required query string, a required knowledgeBaseId string identifying the AWS Knowledge Base, and an optional n number (default 3) controlling how many results to retrieve.
How to install
{
"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"
}
}
}
}
A Docker variant sets command to docker, with args for run, -i, --rm, then a repeated -e flag for each of AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and AWS_REGION before the mcp/aws-kb-retrieval-server image name - passing those three AWS variables through from the host environment rather than hardcoding them in the config's env block. VS Code has one-click install buttons for both the npx and Docker variants, plus a manual path: paste the config into User Settings (JSON) via Ctrl+Shift+P -> Preferences: Open Settings (JSON), or into a shareable .vscode/mcp.json (which omits the wrapping mcp key). The VS Code config style prompts for credentials at connect time instead of storing them in the file, using inputs entries with ids aws_access_key, aws_secret_key, and aws_region, the first two marked password: true so VS Code masks them, referenced in the server's env block as ${input:aws_access_key}, ${input:aws_secret_key}, and ${input:aws_region}. To build the Docker image from source: docker build -t mcp/aws-kb-retrieval -f src/aws-kb-retrieval-server/Dockerfile .
Who it's for
Teams already using an AWS Bedrock Knowledge Base who want an MCP client to query it directly for RAG context, and who are comfortable adopting a reference server from the archived modelcontextprotocol/servers-archived repository rather than an actively maintained standalone project. MIT licensed.
Source README
AWS Knowledge Base Retrieval MCP Server
An MCP server implementation for retrieving information from the AWS Knowledge Base using the Bedrock Agent Runtime.
Features
- RAG (Retrieval-Augmented Generation): Retrieve context from the AWS Knowledge Base based on a query and a Knowledge Base ID.
- Supports multiple results retrieval: Option to retrieve a customizable number of results.
Tools
- retrieve_from_aws_kb
- Perform retrieval operations using the AWS Knowledge Base.
- Inputs:
query(string): The search query for retrieval.knowledgeBaseId(string): The ID of the AWS Knowledge Base.n(number, optional): Number of results to retrieve (default: 3).
Configuration
Setting up AWS Credentials
- Obtain AWS access key ID, secret access key, and region from the AWS Management Console.
- Ensure these credentials have appropriate permissions for Bedrock Agent Runtime operations.
Usage with Claude Desktop
Add this to your claude_desktop_config.json:
Docker
{
"mcpServers": {
"aws-kb-retrieval": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"AWS_ACCESS_KEY_ID",
"-e",
"AWS_SECRET_ACCESS_KEY",
"-e",
"AWS_REGION",
"mcp/aws-kb-retrieval-server"
],
"env": {
"AWS_ACCESS_KEY_ID": "YOUR_ACCESS_KEY_HERE",
"AWS_SECRET_ACCESS_KEY": "YOUR_SECRET_ACCESS_KEY_HERE",
"AWS_REGION": "YOUR_AWS_REGION_HERE"
}
}
}
}
{
"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"
}
}
}
}
Usage with VS Code
For quick installation, use one of the one-click install buttons below...
Manual Installation
For manual installation, add the following JSON block to your User Settings (JSON) file in VS Code. You can do this by pressing Ctrl + Shift + P and typing Preferences: Open Settings (JSON).
Optionally, you can add it to a file called .vscode/mcp.json in your workspace. This will allow you to share the configuration with others.
Note that the
mcpkey is not needed in the.vscode/mcp.jsonfile.
{
"mcp": {
"inputs": [
{
"type": "promptString",
"id": "aws_access_key",
"description": "AWS Access Key ID",
"password": true
},
{
"type": "promptString",
"id": "aws_secret_key",
"description": "AWS Secret Access Key",
"password": true
},
{
"type": "promptString",
"id": "aws_region",
"description": "AWS Region"
}
],
"servers": {
"aws-kb-retrieval": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-aws-kb-retrieval"],
"env": {
"AWS_ACCESS_KEY_ID": "${input:aws_access_key}",
"AWS_SECRET_ACCESS_KEY": "${input:aws_secret_key}",
"AWS_REGION": "${input:aws_region}"
}
}
}
}
}
For Docker installation:
{
"mcp": {
"inputs": [
{
"type": "promptString",
"id": "aws_access_key",
"description": "AWS Access Key ID",
"password": true
},
{
"type": "promptString",
"id": "aws_secret_key",
"description": "AWS Secret Access Key",
"password": true
},
{
"type": "promptString",
"id": "aws_region",
"description": "AWS Region"
}
],
"servers": {
"aws-kb-retrieval": {
"command": "docker",
"args": ["run", "-i", "--rm", "mcp/aws-kb-retrieval-server"],
"env": {
"AWS_ACCESS_KEY_ID": "${input:aws_access_key}",
"AWS_SECRET_ACCESS_KEY": "${input:aws_secret_key}",
"AWS_REGION": "${input:aws_region}"
}
}
}
}
}
Building
Docker:
docker build -t mcp/aws-kb-retrieval -f src/aws-kb-retrieval-server/Dockerfile .
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.