Aurora DSQL MCP Server

1.0.27 verified Fri May 01 auth: no python

An AWS Labs Model Context Protocol (MCP) server for Amazon Aurora DSQL. Provides a Python server that exposes DSQL operations via the MCP interface. Version 1.0.27, actively developed with weekly releases.

pip install awslabs-aurora-dsql-mcp-server

Common errors

error ModuleNotFoundError: No module named 'awslabs_aurora_dsql_mcp_server' ↓

cause Wrong import path: using PyPI name with underscores instead of the actual module name.

fix

Run: pip install awslabs-aurora-dsql-mcp-server then import with from aurora_dsql_mcp_server import AuroraDsqlMcpServer

error TypeError: __init__() got an unexpected keyword argument 'endpoint' ↓

cause The parameter was renamed from 'endpoint' to 'cluster_arn' in v1.0.20.

fix

Use cluster_arn instead of endpoint when creating AuroraDsqlMcpServer.

error botocore.exceptions.NoCredentialsError: Unable to locate credentials ↓

cause AWS credentials not configured. The server needs valid AWS credentials to access DSQL.

fix

Set environment variables AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and optionally AWS_SESSION_TOKEN, or use an IAM role.

Warnings

breaking Python >=3.10 required. Installation on Python 3.9 or earlier will fail. ↓

fix Upgrade Python to 3.10+

gotcha The correct import module is 'aurora_dsql_mcp_server', not 'awslabs_aurora_dsql_mcp_server' — many users mistakenly use the PyPI name with hyphens replaced by underscores. ↓

fix Use `from aurora_dsql_mcp_server import AuroraDsqlMcpServer`

deprecated In v1.0.20+, the constructor argument 'endpoint' was renamed to 'cluster_arn'. Old code using 'endpoint' will raise TypeError. ↓

fix Replace `endpoint=...` with `cluster_arn=...`

Imports

AuroraDsqlMcpServer
wrong
```
from awslabs_aurora_dsql_mcp_server import AuroraDsqlMcpServer
```
correct
```
from aurora_dsql_mcp_server import AuroraDsqlMcpServer
```
PyPI name has hyphens, import uses underscores but the package name on PyPI is 'awslabs-aurora-dsql-mcp-server'. The correct import module is 'aurora_dsql_mcp_server' (without 'awslabs' prefix).

create_app

from aurora_dsql_mcp_server import create_app

Quickstart

Initialize and run the Aurora DSQL MCP server. Requires AWS credentials (via env variables or IAM role) and DSQL cluster/secret ARNs.

from aurora_dsql_mcp_server import AuroraDsqlMcpServer
import os

# Configure AWS credentials via environment variables
server = AuroraDsqlMcpServer(
    region=os.environ.get('AWS_REGION', 'us-east-1'),
    cluster_arn=os.environ.get('DSQL_CLUSTER_ARN', ''),
    secret_arn=os.environ.get('DSQL_SECRET_ARN', '')
)

# Run the MCP server
server.run()