Browse the catalog

PG-MCP is a PostgreSQL Model Context Protocol server that enables AI agents to discover, connect to, query, and understand PostgreSQL databases through a resource-oriented architecture. It extends the reference Postgres MCP implementation with multi-database support, rich catalog information extraction, extension context for PostGIS and pgVector, query execution plan analysis, and robust connection management. The server exposes database schema resources, data access capabilities, and specialized tools for executing read-only SQL queries, all while maintaining security through connection ID abstraction and read-only transaction enforcement. Built with Python 3.13 and asyncpg, it's particularly valuable for developers and data analysts who need AI assistants to interact with PostgreSQL databases without exposing sensitive connection details.

Integrates with AutoCAD through COM automation to enable drawing geometric primitives, adding text objects, sending commands, and reading layouts. Supports floor plan generation, architectural layout design, and CAD workflow automation for design professionals.

Enables LLMs to directly query and interact with Snowflake databases, supporting SELECT queries, schema exploration, and table operations.

MCP server for controlling the kitty terminal emulator. Exposes 40 tools for managing windows, tabs, layouts, text, appearance, and more.

Enables AI agents to retrieve live weather updates for any city via OpenWeatherMap, wrapped in MCP format.

Börsdata unoffcial MCP Server in .Net

Conversation save/load MCP server that provides AI assistants with persistent conversation storage through local file system operations. Built with TypeScript using the Model Context Protocol SDK, fs-extra, and Zod for input validation, the implementation offers two core tools: save-conversation for storing conversation summaries with unique identifiers, and load-conversation for retrieving previously saved content. Features security-hardened path validation to prevent directory traversal attacks, stores files in a dedicated ~/.mcp-conversation-save-load directory, and includes comprehensive error handling with sanitized conversation IDs restricted to alphanumeric characters, hyphens, and underscores. Designed for AI assistants that need to maintain conversation context across sessions, knowledge management workflows, and applications requiring persistent storage of dialogue summaries without external database dependencies.

This MCP server, developed by Hannes Rudolph, provides safe read-only access to SQLite databases for AI assistants. Built with FastMCP, it exposes tools for executing validated SELECT queries, listing tables, and describing table schemas. The implementation stands out with its focus on security, featuring query sanitization, parameter binding, and row limit enforcement. By connecting AI capabilities with local SQLite data sources, this server enables AI systems to explore and analyze structured data while maintaining strict safety controls. It is particularly useful for scenarios requiring AI-assisted data exploration, report generation, or insights extraction from SQLite databases, without risking data integrity or exposing sensitive information.

MCP server giving Claude Desktop / Cursor / Cline ad-free web scraping + search via three tools: fetch_page, fetch_pages_batch, search_web.

Enables AI agents to manage Cloudflare DNS records, including listing, creating, updating, and deleting records for your domains.

Magento MCP server: Claude & ChatGPT manage products, stock, orders and customers in Magento 2 / Adobe Commerce.

An MCP server that fetches real-time documentation for popular libraries like Langchain, Llama-Index, MCP, and OpenAI, allowing LLMs to access updated library information beyond their knowledge cut-off dates.

This MCP server implementation provides a bridge to YouTube's subtitle functionality, enabling AI assistants to download and analyze video subtitles. Developed by setkyar, it offers tools for retrieving video information, listing available subtitle languages, and downloading subtitles in specified languages. The server uses yt-dlp for robust YouTube interaction and is containerized with Docker for easy deployment. It's designed for use cases requiring subtitle analysis, such as content summarization, translation, or accessibility enhancement.

Enables designing and rendering urban street cross-sections from natural language descriptions, and importing real streets from OpenStreetMap.

A production-ready MCP server built with Python, exposing 17 tools to manage users, products, and orders via a JSON-backed REST API, ready for Claude Desktop and any MCP client.

A Docker-based MCP server that provides current weather information for cities worldwide. Enables AI assistants like Claude to fetch temperature, weather conditions, and humidity data through a standardized interface.

Enables interaction with Trello boards, lists, and cards via MCP clients like Claude, allowing read, create, update, and search operations through natural language.

OpenTopography global DEM rasters + point queries (free key)

MCP server integrating VolcEngine's image generation capabilities, enabling text-to-image, image-to-image, and image set generation for AI applications.

This MCP server, developed by kazuph, uses yt-dlp to download subtitles from YouTube videos and connects them to claude.ai via the Model Context Protocol. It supports multiple languages and allows for flexible subtitle retrieval based on user-specified language codes. The implementation is designed for easy integration into Node.js projects and can be run using npx. It's particularly useful for developers looking to incorporate YouTube content analysis or transcription capabilities into their AI-powered applications, enabling natural language queries about video content through Claude.