MCP Db Context

🚀 MCP Server

The MCP (Database Context Provider) Server is designed for large - scale Oracle databases, offering a range of powerful database context functions.

🚀 Quick Start

The MCP Server provides a set of useful tools for interacting with Oracle databases. To start using it, ensure your system meets the requirements and choose an appropriate connection mode.

✨ Features

The MCP Server comes with the following key tools:

• search_columns: Search for table columns containing specific terms.
• get_pl_sql_objects: Retrieve information about all PL/SQL objects matching a specified schema.
• get_object_source: Display the source code of a specific PL/SQL object.
• get_table_constraints: Show all constraints (primary keys, foreign keys, unique constraints, check constraints) on a table.
• get_table_indexes: List all indexes defined on a table for query optimization.
• get_dependent_objects: Find all objects that depend on a specified database object.
• get_user_defined_types: Display information about user - defined types in the database.
• get_related_tables: Show all tables related to a specified table (through foreign key relationships).

🔧 Technical Details

Architecture

The MCP Server uses a three - tier architecture to optimize the performance of large - scale Oracle databases:

1. DatabaseConnector Layer
   • Manages Oracle database connections and query execution.
   • Implements connection pooling and retry logic.
   • Handles raw SQL operations.
2. SchemaManager Layer
   • Implements intelligent schema caching.
   • Provides efficient schema lookup and search capabilities.
   • Manages the cache persisted on disk.
3. DatabaseContext Layer
   • Provides advanced MCP tools and interfaces.
   • Handles authorization and access control.
   • Optimizes schema information for AI use.

Connection Modes

The database connector supports two working modes:

Thin Mode (Default)

By default, the connector uses Oracle's thin mode, a pure Python implementation. This mode:

• Is easier to deploy and set up.
• Is sufficient for most basic database operations.
• Has higher environmental portability.

Thick Mode

For scenarios requiring advanced Oracle features or better performance, you can enable the thick mode:

• Using Docker: Set THICK_MODE = 1 in Docker environment variables.
• Local Installation: Export the THICK_MODE = 1 environment variable and ensure that an Oracle client library compatible with your system architecture and database version is installed.

⚠️ Important Note When using Docker, you don't need to worry about installing Oracle client libraries as they are included in the container (Oracle Instant Client v23.7). The container supports Oracle database versions from 19c to 23ai and is suitable for Linux/arm64 and Linux/amd64 architectures.

📦 Installation

System Requirements

• Python: Version 3.12 or higher (recommended for optimal performance).
• Memory: At least 4GB of available RAM (for large databases with over 10,000 tables).
• Disk Space: A minimum of 500MB of free space for schema caching.
• Oracle Database: Compatible with Oracle Database 11g and higher.
• Network: A stable connection to the Oracle database server.

Performance Considerations

• The initial cache build time may take 5 - 10 minutes (depending on the database size).
• Subsequent startup times are significantly reduced, taking only a few seconds.
• It supports high - concurrency access with excellent performance.

📚 Documentation

Contribution Guide

Community contributions are welcome! Please refer to the project documentation to learn how to participate in development and submit issues.

Support Information

For more information or technical support, please contact project maintainers. The project's license is detailed in the LICENSE file.