Browse the catalog

Local OCR & image analysis via Apple Vision — no cloud, no API keys, ~97% fewer tokens on PDFs.

Enables adversarial collaboration between Claude and GPT for automated code critique, verification, and multi-round debate to improve output quality.

An MCP server that lets restaurants operate their point-of-sale through an LLM agent — edit menus, 86 items, check sales, restock inventory, monitor voids, and more via twenty tools over both stdio and Streamable HTTP transports with per-request tenancy from API keys.

Enables AI assistants to search and retrieve content from the Chinese Minecraft Wiki, including article introductions, full content, and category lists.

An MCP server that provides access to the full FRED API (Federal Reserve Economic Data), including the Maps API for regional and geographic data.

Provides comprehensive filesystem operations, text search/replace, image processing, and Tesseract OCR capabilities for AI agents via the Model Context Protocol.

Auto-exports HawkAPI routes as MCP tools, allowing any MCP-compatible client to call your API.

Vehicle safety data from NHTSA — recalls, complaints, crash ratings, investigations, VIN decoding.

This is an MCP server for tracking your progress through DSA questions with spaced repetition, built-in NeetCode 150 list, and tools to log attempts, get next problems, and view stats.

LibreTranslate MCP — open-source machine translation (BYO endpoint)

An API of Ice and Fire — A Song of Ice and Fire (books/characters/houses)

This Babashka MCP server, developed as an open-source project, enables AI assistants to interact with Babashka, a native Clojure interpreter for scripting. Built with TypeScript and leveraging the @modelcontextprotocol/sdk, it provides tools for executing Babashka code, caching recent command results, and accessing command history through standardized MCP endpoints. The implementation features configurable command timeouts and a roadmap for self-hosting in Babashka itself. By connecting AI capabilities with Babashka's scripting power, this server allows for dynamic code execution and interaction with Clojure ecosystems. It is particularly useful for scenarios involving Clojure-based data processing, system automation, or building AI-assisted development tools for Clojure projects.

MCP server for macOS that enables creating, reading, and searching Apple Notes programmatically through tools like find, get by title or ID, and create.

This MCP server provides Go language server capabilities through gopls integration, enabling AI assistants to perform code navigation and analysis across multiple Go workspaces simultaneously. Built by Gerard Adam using Go 1.24.4 with the Model Context Protocol SDK, it offers 7 specialized tools including go-to-definition, find references, hover information, document symbol listing, workspace symbol search, type definition navigation, and workspace status monitoring. The implementation features multi-workspace architecture with dedicated gopls processes per workspace, robust LSP communication with proper message handling and file management, support for both HTTP and stdio transports, and comprehensive Docker deployment with multi-platform builds, making it valuable for Go development workflows, code exploration across large codebases, and building AI assistants that need deep integration with Go project analysis without manual IDE navigation.

FIWARE MCP Server provides a bridge between AI assistants and FIWARE Context Brokers, enabling interaction with IoT data through natural language. This implementation offers tools for checking broker versions, querying entities, and publishing or updating data in the Context Broker using NGSI-LD format. Built with FastMCP, the server includes robust error handling and timeout management, making it valuable for users who need to monitor and control IoT devices or access contextual data without leaving their AI assistant interface.

Production-grade Quality Engineering framework using Playwright, Model Context Protocol (MCP), and GitHub Copilot AI Agents for hybrid UI & REST API automation.

MCP-Developer-Name is a simple server that returns the current developer's information. Created by Wayne Wei, it uses Node.js with TypeScript and the Model Context Protocol SDK to provide a lightweight service that can be run via npx command or deployed as a Docker container. The implementation allows setting the developer name through an environment variable, making it customizable for different users while maintaining a minimal footprint with just the essential dependencies.

PostFast enables scheduling and publishing social media content to Facebook, Instagram, X, TikTok, LinkedIn, YouTube, BlueSky, Threads, Pinterest, and Telegram. It provides tools for listing connected accounts, creating batch posts with platform-specific controls, uploading media, and retrieving performance analytics. The server integrates with the PostFast social media management platform at postfa.st.

This MCP server provides browser automation capabilities using Puppeteer, enabling AI assistants to interact with web pages through a standardized interface. It offers tools for navigation, taking screenshots, clicking elements, filling forms, and executing JavaScript in a real browser environment. The implementation captures console logs and manages screenshots as resources, making them available to the client. Built with TypeScript and the MCP SDK, it supports both headless operation in Docker containers and visible browser windows when run directly, making it ideal for web testing, data extraction, and automated web interactions without requiring users to write browser automation code.

An Effect-native MCP server that lets local agents silently capture macOS windows, displays, and exact regions, then produce Shottr-style PNGs with programmable annotations, backgrounds, padding, rounded corners, and shadows.