Browse the catalog

A simple MCP server for Claude Desktop with example tools

The Raccoon AI MCP Server enables AI assistants to perform web browsing, data extraction, and complex web task automation through Raccoon's LAM API. Built for Python 3.10+ environments, it allows users to search websites, fill forms, navigate UI elements, extract structured data based on defined schemas, and handle multi-step processes across websites. The server requires a Raccoon AI Secret Key and Passcode from the Raccoon platform, and can be installed via Smithery or from source, making it particularly valuable for workflows requiring web automation and data extraction capabilities without direct browser access.

Enables fetching stock quotes, historical prices as CSV, and plain-text Unicode price charts via Yahoo Finance data.

Precise Symbol-Level Code Retrieval MCP Server: Cut AI Token Costs 95% in 2026

On-demand access to amCharts 5 docs, code examples, and API reference

Swift MCP GUI Server enables AI assistants to control macOS through mouse movements, clicks, and scrolling via the SwiftAutoGUI library. Built by Nakaoka Rei, this server exposes tools for precise cursor positioning, mouse button clicks, and directional scrolling, allowing MCP clients to programmatically interact with the macOS interface. The implementation runs as a native Swift application requiring macOS 15.0+ and Swift 6.0+, making it particularly valuable for automation workflows, GUI testing, or remote control scenarios where programmatic manipulation of the mouse and keyboard is needed.

Dynamic price elasticity of demand calculator deriving marginal revenue maxima

Enables full CRUD operations on calendar events, free/busy queries, and availability slot finding via CalDAV, supporting Nextcloud, Radicale, iCloud, and other RFC 4791-compliant servers.

A template repository for developing Model Context Protocol (MCP) servers with devcontainer setup and curated learning resources.

This tool helps you find the store offering the best price for a specific product among various chain markets in Turkey. The system fetches real-time data from different markets and returns the best price along with the corresponding store name.

Extract data from websites using a fast, intelligent web scraping library designed for modern HTML structures and dynamic content.

Profanity detection and content moderation library.

Manage Mantis Bug Tracker (issues, projects, users, filters, configs) via the official Mantis REST API.

GenPark AI Agent Skill - Runtime parameter type sanitizer, string-to-numeric casting, and boundary constraint validator for LLM tool dispatch.

Exposes Synthetic web search over stdio for local MCP clients. Returns search results with URL, title, publication date, and truncated page text snippets. Sanitizes malformed control characters and limits context size.

This cryptocurrency market data MCP server, developed by an unnamed creator, integrates with major exchanges through the CCXT library to provide real-time and historical crypto data. Built with Python using the Model Context Protocol SDK, it offers tools for fetching current prices, market summaries, trading volumes, and historical OHLCV data across multiple exchanges. The server implements standardized data formatting and robust error handling for exchange connectivity issues. By abstracting the complexities of crypto API interactions, it enables AI systems to easily analyze market trends and access detailed trading information. This implementation is particularly valuable for applications requiring programmatic access to crypto market data, facilitating use cases such as algorithmic trading, portfolio tracking, and market analysis across various cryptocurrencies and exchanges.

Enables users to send live webcam images to Claude Desktop or other MCP clients, facilitating interaction through capturing images, screenshots, and providing a webcam view for visual input.

An MCP for Multipass

X (Twitter) bookmarks MCP server: search, label, and manage your Tweetsmash bookmarks from Claude, ChatGPT, Cursor, and any MCP client