MXCP (Model eXtension Context Protocol) is an enterprise-grade MCP server built specifically for production environments. Unlike simple data access tools, MXCP provides enterprise security, audit logging, policy enforcement, dbt integration, and comprehensive data governance. It transforms any data source into AI-ready endpoints with full compliance tracking and type safety.

You can get MXCP running in under 60 seconds. Simply install via pip (pip install mxcp), create a project (mxcp init --bootstrap), and start serving (mxcp serve). The bootstrap option creates example configurations and connects to Claude Desktop automatically.

Yes, MXCP is fully compatible with Claude Desktop and any other MCP-compliant AI client. It implements the Model Context Protocol standard, ensuring seamless integration with current and future AI tools. Configuration is simple through the Claude Desktop MCP servers config.

MXCP supports a wide range of data sources including PostgreSQL, MySQL, SQLite, DuckDB, CSV files, JSON files, APIs, and cloud storage. It also has native dbt integration for data transformation and caching, making it ideal for complex data pipelines.

MXCP includes enterprise-grade security features including OAuth authentication, role-based access control, policy enforcement, audit logging, data masking, and compliance tracking. Every query is logged with full user context, and policies can enforce who sees what data based on permissions.

MXCP has native dbt integration that creates a powerful data pipeline workflow. dbt models transform and cache data locally in DuckDB, while MCP endpoints query the cached tables directly. This provides lightning-fast AI queries with full data governance and eliminates the need for complex dbt syntax in AI interactions.

MXCP is released under the Business Source License 1.1 (BSL). It is free to use for development, testing, and most production scenarios. However, use of MXCP — or any derivative — as a hosted service or commercial platform that enables third parties to run models, workflows, or database queries (such as a Model Execution Platform, API Marketplace, or DBaaS) requires a commercial license. The license automatically converts to the MIT License four years after the release of each version. You can view the source code and contribute to its development.

Join our Discord community for real-time support, check GitHub Discussions for technical questions, and follow our blog for updates. We also provide comprehensive documentation, examples, and tutorials to help you get the most out of MXCP.