Packages

Provides an “Agent Skills open format” skill containing Core Data best-practice guidance: decision trees and reference documents covering persistence setup, concurrency/thread safety, fetch/performance, schema migration, and CloudKit integration.

go-utcp is a Go implementation of the Universal Tool Calling Protocol (UTCP) that lets clients discover and call tools through multiple transport types (e.g., HTTP, WebSocket, gRPC, GraphQL, TCP/UDP, CLI, SSE/streaming HTTP) and provides utilities like environment-variable substitution (.env) and an OpenAPI-to-manual converter.

Provides a Model Context Protocol (MCP) server for Home Assistant using HTTP transport. It exposes MCP tools/resources/prompts/completions that let an AI assistant query Home Assistant state and call services, and it supports OAuth 2.0 authentication via an external OIDC server (hass-oidc-server), including dynamic client registration for remote use (e.g., Claude in a browser).

Local MCP Gateway is a self-hosted control-plane/proxy that aggregates multiple MCP servers behind profile-specific HTTP endpoints, with a web UI for managing MCP servers, profiles, tools, logging/observability, and access control (API keys and OAuth 2.1).

An MCP (Model Context Protocol) server that provides Kubernetes management tools to AI clients, enabling operations such as applying manifests (including server-side apply), dry-run validation, viewing/scaling workload resources, log streaming, port-forwarding, and multi-cluster context switching.

Provides a Model Context Protocol (MCP) server that exposes MongoDB-backed content to LLMs, with an SSE-based MCP interface for JSON-RPC 2.0 calls and a separate REST CRUD API for managing novels/chapters/characters/Q&A data.

mcp-reasoner is an MCP server implementation for Claude Desktop that provides configurable reasoning strategies—Beam Search and Monte Carlo Tree Search (MCTS)—and reports/evaluates the quality of different reasoning paths. It also includes experimental “alpha” MCTS variants with additional (incomplete/subject-to-change) simulation layers.

An MCP server that helps an agent work with AWS and Terraform-related workflows by exposing tools for tasks such as Terraform configuration/plan/apply and AWS resource interactions (exact tool list not provided in the prompt).

MCP server (mcp-server-fetch) that provides a fetch capability to agents via the Model Context Protocol, enabling them to retrieve external resources as part of tool calls.

Provides an MCP server that exposes Microsoft Graph functionality to MCP-capable AI agents, allowing the agent to perform directory/user/mail/workload actions via Graph-backed tools (exact tool surface depends on the repository implementation).

Converts an MCP server configuration into an Anthropic Claude Skill package that uses progressive disclosure: Claude loads only minimal metadata at startup, loads full tool instructions when the skill is invoked, and executes MCP tool calls externally via a generated executor.

Provides a local read-only MCP server that exposes Microsoft Teams data through CData JDBC Driver-backed, SQL-like tools (get_tables, get_columns, run_query) for AI clients such as Claude Desktop via stdio.

MCP server that exposes Proxmox (virtualization) management capabilities to an AI agent via the Model Context Protocol, enabling tool-based actions against a Proxmox host/cluster.

Radarr is a self-hosted media management application for movies. It lets you add movie releases, searches for matching releases in configured sources, and manages downloading and organization (including renaming, folder placement, and quality profiles) via external download clients.

Provides an MCP server (built with FastMCP) that integrates with RongCloud IM services to support user management, private and group messaging, group lifecycle operations (create/dismiss/join/quit), and retrieval of messages and group members.

A minimal example MCP (Model Context Protocol) server intended for “Deep Research”-style workflows that can search and fetch cupcake order data from a local records.json dataset.

semantic-model-mcp-server is a Python MCP server that connects to Microsoft Fabric and Power BI semantic models to browse workspaces/datasets, retrieve and validate TMSL, execute DAX queries, and create/update models via TMSL. It also includes a Best Practice Analyzer (BPA) that evaluates TMSL against a set of industry/Microsoft-recommended rules, plus tooling for detecting and using local Power BI Desktop instances for development/testing.

Provides an MCP (Model Context Protocol) server that connects an AI agent to Microsoft SQL Server, enabling agent-driven database interactions (e.g., querying/inspecting SQL Server) via MCP tools.

Provides an MCP (Model Context Protocol) server (Python, Docker) that exposes tools for CRUD operations on tables in a Supabase database using SUPABASE_URL and a SUPABASE_SERVICE_KEY.

A Docker-first reference architecture/monorepo for orchestrating multiple AI agents with GitHub Project workflows, tool integration via multiple MCP servers, hardened containerized CI/CD, and research/simulation packages focused on AI agent trust/security concepts (e.g., sleeper-agent detection). It provides Rust CLI tools for board/PR monitoring and validation plus supporting docs and companion packages.