Packages

spacy-server is a service wrapper for spaCy that exposes spaCy NLP capabilities over a network interface (typically as an HTTP API) so other applications can send text and receive NLP annotations.

Provides a demo Spring AI MCP (Model Context Protocol) server that exposes simple tool/function-calling (math: sum/multiply) over SSE, and a client component that connects to that MCP server and offers a REST endpoint (/question) to accept user questions and produce model completions using OpenAI.

An MCP (Model Context Protocol) server package intended to integrate with Obsidian to provide “summary stack” observation/summary-related functionality to AI agents via MCP tools.

A test package (as indicated by its name) providing an MCP server interface for EC2-related capabilities, likely intended for integration testing of “mcp-server” tooling rather than production AWS interaction.

An MCP server (Rust) that controls a physical turret via serial communication, exposing MCP tools to fire, set pan/tilt position, and query bullet count (plus an echo tool).

Provides a Zed Editor context-server integration that connects Zed Assistant to Basic Memory via a MCP server ("basic-memory"). It is configured through Zed settings and optionally selects a Basic Memory project for storing Markdown-based notes.

Provides an MCP (Model Context Protocol) integration for Zed to browse and interact with Buildkite CI/CD resources such as pipelines, builds, jobs/logs, and build artifacts using a Buildkite API token configured in Zed.

zim-mcp-server is described as an MCP (Model Context Protocol) server that likely exposes capabilities related to “zim” (a common ZIM file/wiki format). The provided prompt does not include README/repo contents, so the specific tools, endpoints, and behaviors cannot be verified.

A Flask server implementation of the Simple Todo API defined in `acmetodo.yaml`.

Search and discover Agent Skills from the skills.sh registry. Powered by HAPI MCP server.

[(주)에이아이캐슬](https://aicastle.com)에서 만든 [**OpenAI Agent School** ](https://openai-agent.aicastle.school/)(OpenAI를 활용한 나만의 AI 에이전트 만들기) 강의 자료입니다.

Anet MCP Server is a Rust-based implementation of an MCP (Model Control Protocol) server that exposes JSON-RPC 2.0 methods for tools, prompts, and resources. It uses a NATS transport layer for message passing and an async framework (Tokio) for handling requests.

eureka-server provides an implementation of Netflix-style Eureka service discovery (Eureka Registry) to help microservices register their instances and discover other services through a central registry.

An MCP server (Node.js) that scans a local workspace, builds a semantic index of code, and exposes tools for AI/editor clients to search relevant code in response to natural-language queries. It also maintains an up-to-date index (via file watching) and supports debugging/status/log commands.

IBM FHIR Server is a service that implements the HL7® FHIR standard API for creating, reading, updating, deleting, and searching clinical/health data resources (FHIR resources) over HTTP.

json-server-https is a variant of json-server that serves a JSON file (or JSON-based data) over HTTPS, enabling REST-like CRUD endpoints for local development/testing.

Provides an MCP server that connects an MCP client (e.g., Claude Desktop) to Google Drive, exposing at least one tool to upload Microsoft Word content to a specified Google Drive folder (returning a Google Docs-view URL). Supports both MCP stdio and an SSE/HTTP transport mode, with OAuth2 protection for the SSE/HTTP mode.

mcp-server-aws-cdk is an MCP server package intended to expose AWS CDK-related functionality to an agent via the Model Context Protocol (MCP). It is generally used to help agents scaffold, reason about, or generate AWS CDK constructs/templates through a standardized MCP interface.

Provides a demo Spring AI MCP server using Spring Boot (WebFlux). It exposes an SSE endpoint for streaming MCP communication (e.g., /sse) and an HTTP POST message endpoint (e.g., /mcp/message) to send JSON-RPC requests such as tools/list and notifications/initialized, returning MCP tool definitions and messages over SSE.

A workshop to learn how to build AI-Infused applications with Quarkus and LangChain4j.