DatabaseMCP
An MCP server that enables AI agents to safely explore and interact with MySQL databases through dynamic tool generation from stored procedures. It provides database discovery capabilities and intelligent procedure categorization while enforcing security restrictions to prevent data modification.
README
MCP Database Intelligence Server Architecture
System Overview
The system exposes database capabilities to AI agents using the Model Context Protocol (MCP).
The MCP server acts as a bridge between:
- AI assistants
- database metadata
- stored procedures
- database exploration tools
High Level Architecture
AI Agent / MCP Client │ ▼ MCP Server (FastMCP) │ ├── Discovery Tools │ (database exploration) │ ├── Dynamic Procedure Tools │ (generated from database procedures) │ ▼ Intelligence Engine (procedure metadata + categorization) │ ▼ Database Access Layer │ ▼ MySQL Database
Core Components
1. MCP Server (main.py)
The MCP server is created using FastMCP.
Responsibilities:
- initialize the MCP server
- register discovery tools
- register dynamic procedure tools
2. Database Access Layer (db.py)
This module manages all database communication.
Responsibilities:
- database connection management
- stored procedure execution
- metadata retrieval
- enforcing result limits
All database interactions pass through this layer.
3. Tool Generator (tool_generator.py)
This module dynamically converts database stored procedures into MCP tools.
Steps performed:
- load procedure metadata from database
- detect procedure parameters
- generate MCP tool functions dynamically
- register tools with the MCP server
This allows the system to automatically support new procedures added to the database.
4. Intelligence Engine (intelligence_engine.py)
The intelligence engine analyzes stored procedures to generate metadata.
Capabilities include:
- procedure categorization
- description generation
- parameter analysis
Example categories:
| Procedure Pattern | Category |
|---|---|
| GetLinkCount | Network Analytics |
| GetWeekData | Time Analytics |
| InterfaceLinkType | Interface Analytics |
This metadata helps AI agents understand the purpose of each tool.
5. Discovery Tools (discovery_tools.py)
Discovery tools allow AI agents to explore database structure.
Implemented tools:
list_databases list_tables describe_table
These tools rely on database metadata from:
INFORMATION_SCHEMA
Security Architecture
Several safeguards are implemented to ensure safe database interaction.
Keyword Restrictions
Procedures containing dangerous keywords are blocked.
Examples:
DELETE DROP TRUNCATE UPDATE INSERT ALTER
Query Result Limits
The system limits result size using a configurable value:
MAX_QUERY_ROWS
This prevents large data extraction.
Rate Limiting
The MCP server enforces request rate limits to prevent excessive usage.
Restricted Database User
The database user used by the MCP server should only have:
SELECT EXECUTE
permissions.
This prevents modification of production data.
Metadata Sources
The system retrieves metadata from MySQL system tables.
Examples:
INFORMATION_SCHEMA.ROUTINES INFORMATION_SCHEMA.PARAMETERS INFORMATION_SCHEMA.TABLES INFORMATION_SCHEMA.COLUMNS
These tables allow the MCP server to automatically understand database structure.
Example AI Interaction Flow
Example interaction between an AI agent and the database:
list_databases() ↓ list_tables() ↓ describe_table("interface_stats") ↓ GetLinkCount("NodeA")
This allows the AI to explore database structure before executing analytical procedures.
Extensibility
The system is designed to support additional database objects.
Future extensions may include:
list_views list_functions list_triggers search_tables explain_query
The modular architecture allows new tools to be added easily.
Conclusion
This architecture demonstrates how MCP servers can safely expose database capabilities to AI systems.
Key design principles:
• modular architecture • dynamic tool generation • metadata driven discovery • secure database access
This approach enables safe and scalable AI-database integration.
Detailed system diagrams are available in the docs/ directory.
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.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.