What are the prerequisites?
Node.js v16 or later and pnpm (recommended) are required.
@ralf-boltshauser
基本信息
暂无概览
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"sbb-mcp-server": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://subdomain.yourdomain.com/sse"
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
点击上方按钮,在临时沙箱中运行此服务器并实时获取工具列表。
A starter project for building Model Context Protocol (MCP) servers in TypeScript. It provides a simple echo server implementation that demonstrates core MCP features such as tools, resources, and prompts.
Clone the repository, install dependencies with pnpm install, then run pnpm dev to start the development server and inspector. To switch between STDIO and SSE modes, comment or uncomment the corresponding code blocks in src/index.ts. For STDIO mode, build with pnpm build and run node dist/index.cjs; for SSE mode, deploy using Docker or platforms like Coolify and connect via npx -y mcp-remote https://your-domain.com/sse.
Node.js v16 or later and pnpm (recommended) are required.
STDIO mode is ideal for local development and direct process communication with tools like Claude Desktop. SSE mode is better for production deployments requiring remote access and is used with HTTP/SSE endpoints.
You can deploy using Docker or platforms like Coolify. In Coolify’s advanced settings, disable GZIP compression (required for SSE) and expose port 3001 by setting the domain as `https://your-domain.com:3001`. The server listens on all interfaces by default.
Use the built‑in inspector at `http://localhost:6274`. You can also send debug messages from your code with `server.server.sendLoggingMessage`, which appears in the inspector.
Use a `.env` file for local development (add it to `.gitignore`) or set them in your deployment platform (e.g., Coolify). Access them in code via `process.env`. Docker Compose loads `.env` automatically.

OpenMark is a remote MCP server available at https://openmark.ai. Benchmark your own prompt on 100+ models with real API calls.

Design review for UI code, inside your agent. quick_review checks the files you just changed in about ten seconds. review_files returns a 0-100 score and patches. verify_fixes confirms the fixes landed, free. review_url
Modelglass turns "which model should I use?" into a lookup instead of a guess. Seven tools give your agent live pricing across image, language, video, and audio models. Sourced and versioned, not scraped once and forgott

The tools your AI needs to make your games. Turn prompts into production-ready game assets — animated characters with their weapons attached, seamless textures, terrain, VFX, HUDs — and playable game kits Unity, Godot, U

One case file per task — shared by every session, agent and person that touches it. Self-hosted MCP server and live board. MIT.
评论