# Data interaction

This project designs and implements an architecture that connects the data space with generative AI. Through the Model Context Protocol (MCP) server as a secure middle layer, the language model can indirectly access the DuckDB database. It has the characteristics of modularity, extensibility, and log tracking, laying the foundation for future upgrades to the RAG architecture.

A model context protocol server for interacting with the Foundry platform, supporting AI assistants to operate on datasets, ontology objects, and execute functions.

A lightweight MCP server for seamless integration with Microsoft Dynamics 365 Business Central, supporting interaction with any entity in Business Central via the MCP protocol.

The MCP client tool developed by Pinax is used to connect to services compatible with the MCP protocol, providing real - time data interaction functions based on SSE. Authentication requires obtaining an access token through the thegraph.market platform.

A server that provides access to Lark Bitable through the MCP protocol, supporting SQL queries and operations on table data.

The Databricks MCP Server is a model interaction service based on the MCP protocol, providing access interfaces to functions such as Databricks clusters, jobs, and notebooks, and supporting the integrated interaction between LLM tools and the Databricks platform.

The Salesforce MCP server is a REST API implementation based on jsforce, used to interact with the Salesforce platform, providing functions such as data query, metadata retrieval, record creation, deletion, and modification, and includes a secure authentication mechanism.

The Odoo MCP Server is a middleware that integrates with the Odoo ERP system, allowing AI assistants to access and operate Odoo data and functions through the Model Context Protocol.

A powerful HubSpot CRM integrated MCP server that provides tools for AI assistants to interact with HubSpot data, including contact management, company activity tracking, and real-time data access.

The Databricks MCP Server is a service that implements the Model Completion Protocol (MCP), providing the ability to access Databricks functions via the MCP protocol, enabling LLM tools to interact with Databricks clusters, jobs, notebooks, etc.

An MCP server that connects to the Tana input API to enable data interaction between LLMs and Tana workspaces

A Marvel API MCP server project based on Azure Functions that provides interactive functions for Marvel character and comic data, supports multiple query tools, and integrates with MCP tools (such as VS Code, Claude, etc.).

The Strapi MCP Server is a standardized interface service that enables AI assistants to interact with the Strapi CMS, providing functions such as content type management, REST API operations, and media uploads. It is compatible with Strapi v4 and v5.

The MCP IPFS Server is a Node.js - based server that implements the Model Context Protocol (MCP) for interacting with the storacha.network platform via the w3 command - line tool. It encapsulates the w3 CLI functions, provides operations such as space management, data upload/download, and authorization delegation, and supports language models and other MCP clients to access IPFS storage services through a standardized interface.

The Salesforce MCP Server is an implementation of a model context protocol server that interacts with Salesforce via the REST API. It uses the jsforce library and supports SOQL queries, metadata retrieval, and record operations.

An MCP server that connects Claude to Contentful CMS, supporting the retrieval of content types and entry data.

ngspice-mcp is an MCP protocol-based server that provides a standardized interface for AI language models to access ngspice circuit simulation capabilities, supporting operations such as circuit loading, simulation execution, and data acquisition.

A simple MCP server based on SSE, including sample tools, resources, and prompts, supporting PostgreSQL reading and NATS writing functions.

The Claude desktop client realizes data interaction with Google Cloud by configuring the MCP server

An MCP server that exports pages via the unofficial Notion API, providing a standardized interface for AI assistants to access Notion data.