PostgreSQL Performance MCP

A Model Context Protocol (MCP) server for PostgreSQL database performance analysis and optimization.

Overview

PostgreSQL Performance MCP is a powerful tool that leverages AI to help database administrators and developers optimize their PostgreSQL databases. It provides comprehensive analysis of database structure, query performance, index usage, and configuration settings, along with actionable recommendations for improvement.

This tool runs as a remote MCP server using Server-Sent Events (SSE) transport, allowing it to be deployed centrally and accessed by any MCP-compatible client, including Amazon Q Developer CLI, Claude and other AI assistants that support the MCP protocol.

⚠️ Disclaimer

EXPERIMENTAL: This project is experimental and provided as a demonstration of what's possible with MCP and PostgreSQL. All recommendations and code should be carefully reviewed before implementation in any production environment.

NOT OFFICIAL: This is a personal project and not affiliated with, endorsed by, or representative of any organization I work for or contribute to. All opinions and approaches are my own.

NO LIABILITY: This tool is provided "as is" without warranty of any kind. Use at your own risk. The author is not liable for any damages or issues arising from the use of this software.

Features

- Database Structure Analysis: Analyze tables, columns, indexes, and foreign keys
- Query Performance Analysis: Analyze execution plans and identify bottlenecks
- Index Recommendations: Get suggestions for new indexes based on query patterns
- Query Optimization: Receive suggestions for query rewrites to improve performance
- Slow Query Identification: Find and analyze slow-running queries
- Database Health Dashboard: Get a comprehensive overview of database health metrics
- Index Usage Analysis: Identify unused, duplicate, or bloated indexes
- Read-Only Query Execution: Safely execute read-only queries for verification

Security

This tool operates in read-only mode by default. All database connections are established with SET TRANSACTION READ ONLY to prevent any accidental modifications to your database. The query execution functionality is strictly limited to SELECT, EXPLAIN, and SHOW commands.

Installation

Prerequisites

- Python 3.12+
- Amazon Aurora or RDS PostgreSQL database
- AWS account (for Secrets Manager, optional)

Setup

1. Clone the repository:

   git clone https://github.com/yourusername/postgres-performance-mcp.git

   cd postgres-performance-mcp

   

2. Install dependencies:

   pip install -r requirements.txt

   

3. Configure your database credentials:
- Option 1: Store credentials in AWS Secrets Manager (recommended)
- Option 2: Provide credentials directly when using the tools

Usage

Starting the Server

python server.py --host 0.0.0.0 --port 8000

Configuring MCP Clients

To connect to your remote MCP server, configure your MCP client with:

Server URL: http://your-server-address:8000/
Transport: SSE (Server-Sent Events)

Using with an MCP Client

Connect to the server using any MCP-compatible client and use the available tools:

analyze_database_structure(secret_name="my-postgres-db-credentials", region_name="us-west-2")

Available Tools

- analyze_database_structure: Analyze database schema and provide optimization recommendations
- get_slow_queries: Identify slow-running queries in the database
- analyze_query: Analyze a SQL query and provide optimization recommendations
- recommend_indexes: Recommend indexes for a given SQL query
- suggest_query_rewrite: Suggest optimized rewrites for a SQL query
- database_health_dashboard: Generate a comprehensive health dashboard for the database
- query_optimization_wizard: Interactive wizard to optimize a SQL query step by step
- analyze_index_usage: Analyze index usage patterns and identify unused or inefficient indexes
- execute_read_only_query: Execute a read-only SQL query and return the results
- show_postgresql_settings: Show PostgreSQL configuration settings with optional filtering
- health_check: Check if the server is running and responsive

Customization

The real power of this tool comes from customizing it to your specific environment:

- Add custom analysis rules tailored to your database usage patterns
- Integrate with your monitoring systems
- Customize recommendations based on your organization's best practices
- Add domain-specific knowledge about your data model
- Extend with additional tools specific to your needs

AWS Secrets Manager Setup

To use AWS Secrets Manager for storing database credentials:

1. Create a secret in AWS Secrets Manager with the following keys:
- host: Database hostname
- port: Database port (usually 5432)
- dbname: Database name
- username: Database username
- password: Database password

2. Ensure your AWS credentials are configured with appropriate permissions to access the secret.

3. Use the secret name when calling the tools:

   analyze_query(query="SELECT  FROM users WHERE user_id = 123", secret_name="my-postgres-db-credentials")

   

CREATE EXTENSION pg_stat_statements;
CREATE EXTENSION pg_buffercache;

shared_preload_libraries = 'pg_stat_statements'
pg_stat_statements.track = all

analyze_database_structure(secret_name="my-postgres-db-credentials")

analyze_query(
    query="SELECT  FROM orders JOIN customers ON orders.customer_id = customers.id WHERE orders.status = 'pending'",
    secret_name="my-postgres-db-credentials"
)

Getting Index Recommendations

recommend_indexes(
    query="SELECT * FROM products WHERE category = 'electronics' AND price < 100",
    secret_name="my-postgres-db-credentials"
)

Executing a Read-Only Query

execute_read_only_query(
    query="SELECT schemaname, relname, n_live_tup FROM pg_stat_user_tables ORDER BY n_live_tup DESC LIMIT 10",
    secret_name="my-postgres-db-credentials"
)

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgements

- PostgreSQL community for their excellent documentation
- MCP protocol developers for enabling AI-powered tools