MCP Server using Amazon Q Business Cross App index
About
An MCP server that integrates with Amazon Q Business' cross-app index to answer questions based on uploaded synthetic data. It is designed for users who want to query support tickets and similar documents through Amazon Q Business.
Explore
- Uses Amazon Q Business cross-app index for semantic search.
- Pre‑built sample questions for IT ticket resolution scenarios.
- Managed via uv for Python dependency handling.
- Includes a test client script for validation.
- Compatible with Claude Desktop as an MCP host.
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
MCP Server using Amazon Q Business Cross App indexCommand (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
Create .env file in the root of this module.
touch .env
The application uses environment variables for configuration. You can modify these in the .env file:
REGION=us-east-1
Q_BUSINESS_APP_NAME=REPLACE_WITH_YOUR_Q_BUSINESS_APP_NAME
This project uses uv for dependency management instead of traditional Python venv.
For Mac, you MUST run brew install uv for smooth MCP experience. Otherwise you may run into ENOENT challenges
brew install uv
You can run the server using the provided script:
./run_with_uv.sh
This script will:
1. Create a uv environment if it doesn't exist
2. Install the required dependencies
3. Run the MCP server
If you prefer to set up manually:
```bash
uv venv .uv
source .uv/bin/activate
uv pip install boto3 "mcp[cli]" requests fastmcp httpx python-dotenv
Download and install Claude Desktop for your operating system.
Complete the MCP Host setup by following the For Claude Desktop Users guide.
Turn off Web Search for grounded responses to the data synced.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"mcp server using amazon q business cross app index": {
"amazon-q-index-mcp": {
"command": "uv",
"args": [
"venv",
".uv"
]
}
}
}
}McpServers
{
"amazon-q-index-mcp": {
"command": "uv",
"args": [
"venv",
".uv"
]
}
}This is an MCP (Model Context Provider) server for Acme Company that uses Amazon Q Business' cross-app index.
Prerequisites
Upload Data to S3
Upload the files present in synthetic_data directory to S3 and use this as data source for Q Business APP you create as part of instructions in TVM Set up.
(Token Vending Machine) TVM Set up
Set up the amzn-q-auth-tvm from repository https://github.com/aws-samples/custom-ui-tvm-amazon-q-business
Copy the contents of cdk-outputs.json generated after successful deployment of amzn-q-auth-tvm to tvm_values.json file of this repository. No need to configure QUI.
Configuration
Create .env file in the root of this module.
touch .env
The application uses environment variables for configuration. You can modify these in the .env file:
REGION=us-east-1
Q_BUSINESS_APP_NAME=REPLACE_WITH_YOUR_Q_BUSINESS_APP_NAME
Setup and Running with UV
This project uses uv for dependency management instead of traditional Python venv.
For Mac, you MUST run brew install uv for smooth MCP experience. Otherwise you may run into ENOENT challenges
brew install uv
Running the Server
You can run the server using the provided script:
./run_with_uv.sh
This script will:
1. Create a uv environment if it doesn't exist
2. Install the required dependencies
3. Run the MCP server
Manual Setup
If you prefer to set up manually:
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
