DynamoDB Read-Only MCP
A server that enables LLMs like Claude to query AWS DynamoDB databases through natural language requests, supporting table management, data querying, and schema analysis.
jjikky
README
DynamoDB Read-Only MCP
A server that utilizes the Model Context Protocol (MCP) to query AWS DynamoDB databases. This server allows LLMs like Claude to query DynamoDB data through natural language requests.
Features
This MCP server provides the following features:
- Table Management Tools:
list-tables
: View a list of all DynamoDB tablesdescribe-table
: View detailed information about a specific table
- Data Query Tools:
scan-table
: Scan all or part of a table's dataquery-table
: Search for data that matches specific conditions in a tablepaginate-query-table
: Retrieve data across multiple pages that matches specific conditionsget-item
: Retrieve an item with a specific keycount-items
: Calculate the number of items in a table
- Resources:
dynamodb-tables-info
: A resource that provides metadata for all tablesdynamodb-table-schema
: A resource that provides schema information for a specific table
- Prompts:
dynamodb-query-help
: A help prompt for writing DynamoDB queries
Installation and Execution
You can run it without installation using the Run with NPX
method below.
Installation
-
Clone the repository:
git clone https://github.com/jjikky/dynamo-readonly-mcp.git cd dynamo-readonly-mcp
-
Install the required packages:
npm install
-
Create a
.env
file and set up your AWS credentials:AWS_ACCESS_KEY_ID=your_access_key AWS_SECRET_ACCESS_KEY=your_secret_key AWS_REGION=your_region
Build and Run
npm run build
npm start
Connect to Claude Desktop
To use this MCP server with Claude Desktop, you need to modify the Claude Desktop configuration file.
-
Open the Claude Desktop configuration file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json
- Windows:
%APPDATA%\Claude\claude_desktop_config.json
- macOS:
-
Add the server configuration as follows:
{ "mcpServers": { "dynamodb-readonly": { "command": "node", "args": ["/absolute-path/dynamo-readonly-mcp/dist/index.js"], "env": { "AWS_ACCESS_KEY_ID": "your_access_key", "AWS_SECRET_ACCESS_KEY": "your_secret_key", "AWS_REGION": "your_region" } } } }
-
Restart Claude Desktop.
Run with NPX
You can also run this server using npx
without a global installation:
{
"mcpServers": {
"dynamodb-readonly": {
"command": "npx",
"args": ["-y", "dynamo-readonly-mcp"],
"env": {
"AWS_ACCESS_KEY_ID": "your_access_key",
"AWS_SECRET_ACCESS_KEY": "your_secret_key",
"AWS_REGION": "your_region"
}
}
}
}
Usage Examples
You can ask Claude questions like:
- "Can you tell me what tables are in DynamoDB?"
- "Explain the structure of the Users table"
- "Find the number of users in the 'Users' table where groupId is '0lxp4paxk7'"
Architecture
This MCP server consists of the following layered structure:
- Client Interface (Claude Desktop) - Interaction between user and LLM
- MCP Protocol Layer - Provides standardized message exchange method
- DynamoDB Server - Implements functions that interact with DynamoDB
- AWS SDK - Communicates with AWS DynamoDB service
Key Operation Mechanisms
1. Initialization and Connection
When the server starts, the following process occurs:
async function main() {
const transport = new StdioServerTransport();
await server.connect(transport);
console.error('DynamoDB read-only MCP server is running...');
}
StdioServerTransport
sets up a communication channel through standard input/output.server.connect(transport)
connects to Claude Desktop through the MCP protocol.- During connection, the server sends information about supported tools, resources, and prompts to the client.
2. Tool Request Processing
When a user asks Claude something like "Show me the list of DynamoDB tables":
- Claude analyzes this request and calls the
list-tables
tool. - This request is sent to the server through the MCP protocol.
- The server executes the corresponding tool handler:
server.tool('list-tables', 'Gets a list of all DynamoDB tables', {}, async () => {
try {
const tables = await listTables();
return {
content: [{ type: 'text', text: JSON.stringify(tables, null, 2) }],
};
} catch (error) {
return { isError: true, content: [{ type: 'text', text: `Error: ${error.message}` }] };
}
});
- The result is returned to Claude through the MCP protocol.
- Claude processes this result into natural language and presents it to the user.
3. Specific Parameter Handling
When a user requests "Tell me the structure of the Users table":
- Claude determines that this request should use the
describe-table
tool. - Claude configures the parameter as
{ tableName: "Users" }
. - This information is sent to the MCP server:
server.tool(
'describe-table',
'Gets detailed information about a DynamoDB table',
{
tableName: z.string().describe('Name of the table to get detailed information for'),
},
async ({ tableName }) => {
// Query table information using the tableName parameter
const tableInfo = await describeTable(tableName);
// Return results
}
);
Here, z.string()
uses the Zod library to validate parameters.
4. Resource Handling
Resources are another MCP feature that provides read-only data:
server.resource('dynamodb-tables-info', 'DynamoDB table information', async () => {
// Create and return resource data
const tables = await listTables();
const tablesInfo = await Promise.all(/* Query table information */);
return {
contents: [
{
uri: 'dynamodb://tables-info',
text: JSON.stringify(tablesInfo, null, 2),
mimeType: 'application/json',
},
],
};
});
Claude accesses resources and uses them as context information.
5. Prompt Handling
The MCP server can provide prompt templates for specific tasks:
server.prompt(
'dynamodb-query-help',
'A prompt that helps write DynamoDB queries',
{
tableName: z.string().describe('Table name to query'),
queryType: z.enum(['basic', 'advanced']).default('basic'),
},
async ({ tableName, queryType }) => {
// Generate prompt content
return {
messages: [
{
role: 'user',
content: { type: 'text', text: helpContent },
},
],
};
}
);
This prompt is used when a user requests "Show me how to write queries for the Users table."
Data Flow Summary
- User makes a request to Claude in natural language
- Claude analyzes the request and selects the appropriate MCP tool/resource/prompt
- MCP client sends the request to the server in a standardized format
- Server processes the request and calls the AWS DynamoDB API
- DynamoDB returns results
- Server converts results to MCP format and sends them to the client
- Claude processes the results into natural language and presents them to the user
License
This project is licensed under the MIT License - see the LICENSE file for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
MCP Package Docs Server
Facilitates LLMs to efficiently access and fetch structured documentation for packages in Go, Python, and NPM, enhancing software development with multi-language support and performance optimization.
Claude Code MCP
An implementation of Claude Code as a Model Context Protocol server that enables using Claude's software engineering capabilities (code generation, editing, reviewing, and file operations) through the standardized MCP interface.
@kazuph/mcp-taskmanager
Model Context Protocol server for Task Management. This allows Claude Desktop (or any MCP client) to manage and execute tasks in a queue-based system.
Linear MCP Server
Enables interaction with Linear's API for managing issues, teams, and projects programmatically through the Model Context Protocol.
mermaid-mcp-server
A Model Context Protocol (MCP) server that converts Mermaid diagrams to PNG images.
Jira-Context-MCP
MCP server to provide Jira Tickets information to AI coding agents like Cursor

Linear MCP Server
A Model Context Protocol server that integrates with Linear's issue tracking system, allowing LLMs to create, update, search, and comment on Linear issues through natural language interactions.

Sequential Thinking MCP Server
This server facilitates structured problem-solving by breaking down complex issues into sequential steps, supporting revisions, and enabling multiple solution paths through full MCP integration.