AWS Knowledge Base MCP Server
Description
# AWS Knowledge Base MCP Server ## Usage <table><tr><td>Claude Desktop</td><td>GitHub Copilot</td></tr><tr><td> ```json { "mcpServers": { "knowledge-base": { "command": "npx", "args": [ "mcp-remote", "https://aws-knowledge-base-mcp-server.daohoangson.workers.dev/sse" ] } } } ```…
About
# AWS Knowledge Base MCP Server ## Usage <table><tr><td>Claude Desktop</td><td>GitHub Copilot</td></tr><tr><td> ```json { "mcpServers": { "knowledge-base": { "command": "npx", "args": [ "mcp-remote", "https://aws-knowledge-base-mcp-server.daohoangson.workers.dev/sse" ] } } } ``` </td><td> ```json { "servers": {…
Details
- Author
- daohoangson
- GitHub stars
- 1
- Downloads
- 278
- Categories
- Knowledge Base
Jump to
- Provides a search_knowledge_base tool for AI assistants.
- Integrates with AWS Bedrock Knowledge Base for document retrieval.
- Supports both SSE and streamable HTTP transports.
- Runs on Cloudflare Workers with Node.js compatibility.
- Includes infrastructure as code via AWS CDK.
- Comes with a demo client using AI SDK by Vercel.
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
AWS Knowledge Base MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Configure it in Claude Desktop by adding a JSON entry to mcpServers with npx mcp-remote pointing to https://aws-knowledge-base-mcp-server.daohoangson.workers.dev/sse. For GitHub Copilot, add a servers entry with type: "sse" and the same URL. The client demo can connect to the server using SSE or streamable HTTP transports via npm start with the appropriate URL.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"aws knowledge base mcp server": {
"aws-knowledge-base-mcp-server": {
"command": "npx",
"args": [
"cdk",
"deploy"
]
}
}
}
}McpServers
{
"aws-knowledge-base-mcp-server": {
"command": "npx",
"args": [
"cdk",
"deploy"
]
}
}AWS Knowledge Base MCP Server
Usage
<table><tr><td>Claude Desktop</td><td>GitHub Copilot</td></tr><tr><td>
{
"mcpServers": {
"knowledge-base": {
"command": "npx",
"args": [
"mcp-remote",
"https://aws-knowledge-base-mcp-server.daohoangson.workers.dev/sse"
]
}
}
}
</td><td>
{
"servers": {
"aws-knowledge-base": {
"type": "sse",
"url": "https://aws-knowledge-base-mcp-server.daohoangson.workers.dev/sse"
}
}
}
</tr><tr><td>
</td><td>
</td></tr></table>
A Model Context Protocol (MCP) server implementation that enables AI assistants to search through Knowledge Base using AWS Bedrock and Cloudflare Workers. This project consists of three main components:
Architecture
1. Infrastructure (/cdk): Sets up the resources including:
- AWS Bedrock Knowledge Base for document embeddings
- Pinecone for efficient document search
- AWS S3 bucket for storing documentation files
- AWS Lambda function for MCP server with streamable HTTP transport
- AWS IAM user and policies for API access
2. MCP Server (/cloudflare-mcp-server): Implements the MCP server that:
- Provides a search_knowledge_base tool for AI assistants
- Integrates with AWS Bedrock for document retrieval
- Supports both SSE (deprecated) and streamable HTTP transports
- Runs on Cloudflare Workers with Node.js compatibility
3. MCP Client (/ai-sdk-mcp-client): A demo client that:
- Uses AI SDK by Vercel with Nova Pro model
- Acts as a MCP client, supporting both SSE and streamable HTTP transports
- Streams text responses with tool calls
graph TD
subgraph AWS
S3[S3 Bucket] -->|Stores Documents| KnowledgeBase
TitanModel[Titan Embed Text V2] -->|Embedding Model| KnowledgeBase
Lambda -->|Retrieve API| KnowledgeBase
NovaPro[Nova Pro]
end
subgraph Pinecone
KnowledgeBase[Bedrock Knowledge Base] -->|Vector Embeddings| PineconeDb[Vector Database]
end
subgraph Cloudflare
Worker -->|Retrieve API| KnowledgeBase
end
Demo -->|InvokeModel API| NovaPro
Demo -->|MCP<br />Streamable HTTP| Lambda
Demo -->|MCP<br />Server Sent Event| Worker
Demo -->|MCP<br />Streamable HTTP| Worker
Infrastructure
Setup
```bash
cd cdk
npm install
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
