localdata
A dynamic MCP server for local databases and text files with enhanced security and query buffering capabilities
claude mcp add --transport stdio chrisgve-localdata-mcp localdata-mcp
How to use
LocalData MCP Server provides a comprehensive interface for working with databases, spreadsheets, and structured data files with built-in security and performance features. Once started, you can connect to a variety of data sources through the provided tools, enabling you to establish connections, query data, and inspect schemas in a unified environment. The available tools include connect_database for establishing connections to SQL databases, spreadsheets, and file-based sources; disconnect_database to close active connections; list_databases to view current active connections; execute_query to run SQL queries with automatic chunking for large results; next_chunk to fetch subsequent portions of large results; describe_database and describe_table to inspect schemas; and find_table to locate tables by name. These tools are designed to simplify working with diverse data sources while providing structured results and helpful error messages.
How to install
Prerequisites:
- Python 3.10+ installed on your system
- Internet access to fetch packages
Install the MCP server package:
pip install localdata-mcp
Alternatively, install via uv (recommended for quick start):
uv tool install localdata-mcp
For development or local testing:
git clone https://github.com/ChrisGVE/localdata-mcp.git
cd localdata-mcp
pip install -e .
Configuration note: The MCP server is typically run through the mcp_config descriptor. The example in the project configuration uses a single server named localdata. Ensure your environment satisfies any prerequisites for your chosen data sources and that network access is permitted to those sources.
Additional notes
Tips and notes:
- Environment variables: The example configuration includes an empty env block. Add any required environment variables such as database connection strings, credentials, or feature flags as needed, e.g., "env": { "DB_PASSWORD": "your-password" }.
- Docker usage: If you prefer containerized deployment, refer to the Docker Usage Guide mentioned in the README for container deployment and configuration details.
- Security: LocalData enforces path security, SQL validation, and connection limits. Configure per-database timeouts and memory controls to prevent resource exhaustion.
- Upgrading: If migrating between versions, consult the Migration Guide referenced in the release notes for step-by-step migration instructions.
- Troubleshooting: Enable detailed JSON logging to inspect query metrics and security events for easier debugging.
Related MCP Servers
python-notebook
Lightweight Python Notebook MCP - Enable AI assistants to create, edit, and view Jupyter notebooks via Model Context Protocol
fastmcp-builder
A comprehensive Claude Code skill for building production-ready MCP servers using FastMCP. Includes reference guides, runnable examples, and a complete implementation with OAuth, testing, and best practices.
postgresql-ssh
PostgreSQL MCP server with SSH tunneling for Claude Desktop and ChatGPT
cc-session-search
MCP server for searching and analyzing Claude Code conversation history
mcp-tidy
CLI tool to visualize and manage MCP server configurations in Claude Code. List servers, analyze usage statistics, and clean up unused servers
mcpman
The package manager for MCP servers — install, manage & monitor across Claude Desktop, Cursor, VS Code, Windsurf