aws-kb

A Model Context Protocol (MCP) server that enables querying AWS Bedrock Knowledge Bases. This server allows AI assistants to access and query your AWS Bedrock Knowledge Base through a standardized interface.

Features

• Query AWS Bedrock Knowledge Bases using natural language
• Support for context-aware queries
• Configurable response generation parameters
• Easy integration with MCP-compatible AI assistants

Installation

# Install globally
npm install -g aws-kb

# Or install as a project dependency
npm install aws-kb

Prerequisites

• Node.js >= 18.0.0
• An AWS account with Bedrock access
• A configured AWS Bedrock Knowledge Base
• AWS credentials with appropriate permissions

Configuration

The server requires the following environment variables to be set:

Variable	Description
AWS_ACCESS_KEY_ID	Your AWS access key
AWS_SECRET_ACCESS_KEY	Your AWS secret key
AWS_REGION	The AWS region where your Knowledge Base is located
AWS_KNOWLEDGE_BASE_ID	The ID of your Knowledge Base
AWS_INFERENCE_PROFILE_ARN	The ARN of the inference profile to use

You can set these environment variables in your shell:

export AWS_ACCESS_KEY_ID="your_access_key"
export AWS_SECRET_ACCESS_KEY="your_secret_key"
export AWS_REGION="your_region"
export AWS_KNOWLEDGE_BASE_ID="your_kb_id"
export AWS_INFERENCE_PROFILE_ARN="your_profile_arn"

Or create a .env file in your project:

AWS_ACCESS_KEY_ID=your_access_key
AWS_SECRET_ACCESS_KEY=your_secret_key
AWS_REGION=your_region
AWS_KNOWLEDGE_BASE_ID=your_kb_id
AWS_INFERENCE_PROFILE_ARN=your_profile_arn

Usage

As a Command Line Tool

When installed globally, you can run the server directly:

aws-kb

As a Module

import { AWSKnowledgeBaseServer } from 'aws-kb';

const server = new AWSKnowledgeBaseServer();
server.run().catch(console.error);

Using with Claude

1. Add the server configuration to your Claude settings:

{
  "mcpServers": {
    "aws-kb": {
      "command": "aws-kb",
      "env": {
        "AWS_ACCESS_KEY_ID": "your_access_key",
        "AWS_SECRET_ACCESS_KEY": "your_secret_key",
        "AWS_REGION": "your_region",
        "AWS_KNOWLEDGE_BASE_ID": "your_kb_id",
        "AWS_INFERENCE_PROFILE_ARN": "your_profile_arn"
      }
    }
  }
}

2. The server provides a query_knowledge_base tool that accepts the following parameters:

Parameter	Type	Required	Description
query	string	Yes	The question or query to ask
context	string	No	Additional context to help inform the query
temperature	number	No	Temperature parameter for response generation (0.0 to 1.0)

Example Usage

// Example query to send to the server
const query = {
  jsonrpc: '2.0',
  id: 1,
  method: 'callTool',
  params: {
    name: 'query_knowledge_base',
    arguments: {
      query: 'What is AWS Bedrock?',
      context: 'I am new to AWS services',
      temperature: 0.7
    }
  }
};

// Send the query to the server
server.stdin.write(JSON.stringify(query) + '\n');

For more examples, check out the examples directory.

Development

To build from source:

git clone https://github.com/assaads/aws-kb.git
cd aws-kb
npm install
npm run build

Testing

You can use the MCP Inspector to test the server locally:

npm run inspector

Scripts

• npm run build - Build the TypeScript source
• npm run watch - Watch for changes and rebuild
• npm start - Run the server
• npm run inspector - Run the MCP inspector for testing

API Reference

Server Configuration

The server can be configured through environment variables and constructor options:

interface ServerConfig {
  name?: string;
  version?: string;
  capabilities?: {
    tools?: Record<string, unknown>;
  };
}

const server = new AWSKnowledgeBaseServer({
  name: 'custom-name',
  version: '1.0.0'
});

Query Parameters

interface QueryParams {
  query: string;
  context?: string;
  temperature?: number;
}

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Related Projects

• Model Context Protocol
• Claude
• AWS Bedrock