Packages

Provides a Spring Boot MCP server that exposes PTC ThingWorx entities (Things, Properties, Services, and Collections) as AI-friendly tools/endpoints. It uses ThingWorx REST API access with AppKey authentication and includes Redis-backed caching with runtime toggles, configurable via environment variables, and Docker support.

A packaged “agent skill” that helps AI coding agents generate and maintain AGENTS.md files (and optional scoped subsystem variants) according to the agents.md specification. It includes scripts/templates for detecting project structure and stack signals and then producing/updating the agent context files idempotently with managed headers.

An MCP server that enables agents to query Amazon CloudWatch Application Signals (application-level metrics/traces) through MCP tools/actions.

Provides a Python-based Model Context Protocol (MCP) server exposing tools to scrape social media profile data (LinkedIn, Facebook, Instagram) and perform Google search via Serper/Google search. It is intended to be run locally (e.g., with Claude Desktop) and uses external APIs (RapidAPI and Serper) authenticated via environment variables.

EasyR1 is an open-source RL training framework (a fork of veRL) for efficient, scalable reinforcement learning (e.g., GRPO/DAPO/Reinforce++/ReMax/RLOO/GSPO/CISPO) on text and vision-language (multi-modality) models, leveraging vLLM (SPMD mode), FlashAttention, and Ray for multi-node scaling. It provides training scripts, dataset formatting guidance, and checkpoint utilities for model merging/export.

HXAudioPlayer is an Android Java library that wraps/augments MediaPlayer and SoundPool for music and sound-effect playback. It provides a fluent API (HXMusic, HXSound) for loading from resources or URLs/paths, controlling playback (play/pause/resume/stop), and managing sound engines on different Android versions, including notes for gapless playback and SoundPool memory workarounds.

MCP Config Manager is a local CLI/GUI utility for managing Model Context Protocol (MCP) server configuration files across multiple client systems (Claude, Gemini, Codex, etc.). It enables/disables servers, edits server configuration fields, performs backups/restores, supports preset management, and can discover servers from project configuration files.

mcp-foundry is an MCP (Model Context Protocol) server (Python/stdio) intended to provide a unified set of tools for Azure AI Foundry, covering model discovery/details, deployment-related operations, knowledge/index management for Azure AI Search, evaluation utilities for text/agents, and fine-tuning/job status and related management. The README indicates this repo is a deprecated/older experimental implementation and that a newer Foundry MCP Server exists as a cloud-hosted preview.

A gateway that translates MCP traffic between Claude Desktop’s stdio-based MCP client and HTTP/SSE-based MCP servers by accepting stdio requests, forwarding them to an SSE MCP server, and converting the responses back to the stdio MCP format.

MCP server that exposes a filesystem interface to an AI agent, enabling read/write (and possibly directory listing/search) operations through MCP tools, based on a server-side configuration of allowed paths/permissions.

An MCP (Model Context Protocol) server that provides read-only access to the Telegram API (via MTProto/Telethon), exposing tools for listing dialogs (chats/channels/groups) and fetching unread messages from a given dialog.

This repository provides a framework and set of specialized skills for AI coding agents to standardize and automate software engineering tasks.

Provides an MCP (Model Context Protocol) server that exposes MySQL capabilities to an AI agent, enabling the agent to connect to a MySQL database and run SQL queries via MCP tools.

MCP server for Oracle OCI Fleet Application Management (FAM), enabling AI agents to manage fleets of applications, perform patch management, compliance checking, and lifecycle operations across multiple OCI resources and environments.

An MCP server intended to help agents work with PDF documents (e.g., converting/extracting/processing PDFs) via Model Context Protocol tools.

Plan Cascade is an AI-assisted, multi-stage software development framework that decomposes a project goal into a mega plan, per-feature PRDs/design docs, parallel stories/tasks, and “quality gates” (verification, lint/format, test checks, and review). It supports multiple execution modes including a Claude Code plugin, a CLI, and a desktop app, and includes an MCP server component (FastMCP) plus an MCP client/client-side integration for tools.

The package provides an MCP (Model Context Protocol) server that exposes Gmail-related actions/tools to AI agents, enabling programmatic access to a user’s Gmail for tasks like reading and managing email.

MCP (Model Context Protocol) server package intended to provide “scholar”/academic research capabilities to an MCP-capable agent. Provides a programmatic interface for research-oriented tools (exact endpoints/tools not provided in the prompt content).

Evaluates source artifacts for security issues via an MCP (Model Context Protocol) server interface, intended to be used by AI agents to run or orchestrate scanning workflows.

An AI agent skill that acts as your expert Shopify App Store release consultant — guiding you interactively through every deliverable needed to submit and list your app.