Packages

polygon-mcp-server is an MCP (Model Context Protocol) server that exposes Polygon blockchain-related capabilities to MCP-compatible AI agents, allowing them to query/act on Polygon data via MCP tools.

A Claude Code skill that generates responsive, cross-client HTML email templates by producing MJML (MJML 4.x) and compiling it to production-ready, minified HTML suitable for Outlook and Gmail, including Outlook-specific handling (VML/ghost tables), Gmail clipping prevention, and accessibility-oriented MJML structuring.

MCP (Model Context Protocol) server that exposes Elasticsearch operations (via Starsky iFlow integration) to LLM/AI agents through MCP tools, enabling querying and indexing/search workflows against an Elasticsearch backend.

toolhive-cloud-ui is a Next.js web UI that lets you visualize MCP (Model Context Protocol) servers running in your infrastructure, surfacing metadata and providing copy-ready endpoints to speed up AI agent integrations. It relies on toolhive-registry-server as a backend implementing the MCP Registry API and uses OIDC-based authentication (with an OIDC mock and MSW mock API for local development).

TriageMCP is an MCP server intended to perform basic static triage of Windows PE (Portable Executable) files using tooling such as pefile and YARA, returning analysis results to an LLM-driven workflow.

zellij-mcp-server is an MCP server that exposes Zellij (terminal multiplexer) functionality as MCP tools/operations so AI agents can interact with a user's Zellij session/workspace via the MCP protocol.

Provides an Axum-based MCP server (compiled to WebAssembly in the README) that exposes an HTTP JSON-RPC endpoint (/api/counter in the example) and implements at least a sample tool named "counter" with operations like increment, decrement, and get_value.

Brave ("brave") is a browser-oriented package/tooling that implements Brave Search integration. It provides an API client/adapter layer for searching the web using Brave Search, intended for programmatic use by applications and agents.

A Shopify-integrated, AI-powered commerce chatbot that provides customer support and product discovery using hybrid retrieval (FAISS + Elasticsearch) and can perform agentic shopping actions (e.g., create/update cart and customer-related actions) through Shopify GraphQL APIs, with session persistence via Redis and chat history via MongoDB. Includes a Shopify Theme Extension for UI embedding and a FastAPI backend with Docker-based deployment. Mentions an MCP server in the architecture diagram but provides no MCP server endpoint or tooling details in the README.

A Claude Code “skill” that guides systematic genealogy research using Genealogical Proof Standard (GPS): analyzes historical documents (including handwritten, multi-language), helps manage an Obsidian vault (People/Places/Documents/Events, with cross-references and PROCESS.md), recommends sources/databases by time/region, handles naming/language variations, and supports GEDCOM import/export.

GLaDOS is an on-device (Python) voice assistant/agent framework that combines speech recognition, voice activity detection, text-to-speech, vision processing, an LLM core, and an MCP-based tool system to enable proactive/autonomous behavior (e.g., responding to camera/audio/time events) with long-term memory and configurable LLM backends (e.g., Ollama or OpenAI-compatible APIs).

HasMCP-CE (HasMCP Community Edition) converts existing API endpoints described by OpenAPI/Swagger (and/or manually selected endpoints) into an MCP (Model Context Protocol) server that can be run self-hosted. It provides OAuth2 authentication options, endpoint toggling per MCP server, optional proxy header handling to the upstream API, token management, and logging/analytics for MCP tool calls.

mail-mcp-server is a Java/Spring AI MCP Server example that exposes email-related capabilities as MCP tools (notably: find contact email by name, and send an email via Spring Boot’s mail integration). It is intended to be used by an MCP Host/Client (e.g., via Spring AI’s MCP client) to let an LLM trigger these tools.

Provides an MCP (Model Context Protocol) server setup for Graphiti-based knowledge graph extraction and storage in Neo4j. It adds a CLI that generates a root MCP server plus multiple project-specific MCP servers (via Docker Compose) so multiple knowledge graphs can share the same Neo4j database while staying isolated by project configuration.

parse-server is a Node.js server that implements the Parse API spec (Parse backend) so you can build Parse-style applications. It persists data via supported database adapters (e.g., MongoDB) and supports Parse features like classes/objects, queries, file storage, push, and Cloud Code (depending on configuration).

prometheus-server is a Prometheus server component that scrapes metrics and provides time-series storage/querying and alerting support when paired with the broader Prometheus ecosystem.

An MCP server package intended to let an AI agent access Strava activity data by exposing MCP tools/endpoints for querying and retrieving Strava activities (e.g., listing activities, fetching activity details), likely using Strava’s API and an OAuth-based connection.

Unity-focused template for creating AI Model Context Protocol (MCP) tools that integrate with Unity Editor/Runtime, including scaffolding for an MCP plugin tool surface (via annotated C# methods) and helper scripts for packaging/versioning/deployment.

MCP server wrapper for Wolfram|Alpha, exposing Wolfram|Alpha capabilities as MCP tools so AI agents can query for mathematical/scientific answers via the MCP protocol.

Provides an MCP server that exposes WorkOS GitHub-related functionality to AI agents via the Model Context Protocol, enabling tools/actions against GitHub (as supported by the underlying WorkOS GitHub integration).