Redshift Utils

🤖 Enable AI assistants (Claude, Cursor) to monitor, diagnose, and query Amazon Redshift using this MCP server and the AWS Data API.

python

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "vinodismyname-redshift-utils-mcp": {
      "command": "uvx",
      "args": [
        "redshift-utils-mcp"
      ],
      "env": {
        "AWS_REGION": "us-east-1",
        "AWS_PROFILE": "my-redshift-profile",
        "REDSHIFT_DATABASE": "your_database_name",
        "REDSHIFT_CLUSTER_ID": "your-cluster-id",
        "REDSHIFT_SECRET_ARN": "arn:aws:secretsmanager:us-east-1:123456789012:secret:your-redshift-secret-XXXXXX"
      }
    }
  }
}

You deploy the Redshift Utils MCP Server to securely connect your LLMs or AI assistants to your Amazon Redshift data using the Model Context Protocol. It enables natural-language style querying, schema discovery, table inspection, health checks, and performance analysis against your Redshift data via the Data API, all in a standardized MCP interface suitable for Claude, Cursor, or custom clients.

How to use

Use this MCP server to access Redshift data through compatible MCP clients. You can discover schemas, list tables, inspect table metadata, run read-only queries, diagnose performance, monitor workloads, retrieve table definitions, and check cluster health via predefined tools.

Typical usage patterns include launching the MCP server locally and connecting your MCP client (such as Claude Desktop or Cursor IDE) to the server. You’ll interact with the server through MCP resources and tools to perform data exploration and diagnostic tasks without embedding credentials in your prompts.

How to install

Prerequisites you need before installing the MCP server are: Python 3.10 or newer and a local or containerized runtime for running Python-based services. You will also need access to an Amazon Redshift cluster and AWS credentials with Redshift Data API permissions and Secrets Manager access.

Install from PyPI (recommended) and start the server by running the MCP entry point.

Install from source if you prefer to build from the repository.

# Install from PyPI
pip install redshift-utils-mcp

# Run the MCP server (using uvx as the launcher)
uv pip install redshift-utils-mcp

# If you prefer not to install, you can run the server directly via uvx after cloning the repo
# (see the source for exact repository commands and file structure)

Available tools

handle_check_cluster_health

Performs a health assessment of the Redshift cluster using diagnostic SQL scripts to surface health indicators and potential issues.

handle_diagnose_locks

Identifies active lock contention and blocking sessions to help diagnose performance bottlenecks.

handle_diagnose_query_performance

Analyzes a specific query's execution performance, including plan, metrics, and historical data for a given query_id.

handle_execute_ad_hoc_query

Executes an arbitrary SELECT query via the Redshift Data API to retrieve data based on user prompts.

handle_get_table_definition

Retrieves the DDL (SHOW TABLE) output for a specified table, providing its definition.

handle_inspect_table

Fetches detailed information about a table’s design, storage, health, and usage.

handle_monitor_workload

Analyzes cluster workload patterns over a time window, including top queries and resource usage.

Built by

VeilStrat

AI signals for GTM teams