What runtime do I need to run EZ-MCP templates?
Python templates require `uv` (install via `curl -LsSf https://astral.sh/uv/install.sh | sh`). TypeScript templates require Deno (install via `curl -fsSL https://deno.land/install.sh | sh`).
@intellectronica
基本信息
The easiest path to getting an MCP server going
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"ez-mcp": {
"command": "uv",
"args": [
"run",
"ez-mcp.py"
]
}
}
}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。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
EZ-MCP is a collection of ready-to-run, single-file MCP server templates in Python/uv and TypeScript/Deno. It provides instant setup with all core MCP features (resources, tools, prompts) and is designed for developers who want to build or experiment with MCP servers in minutes.
Clone the repository, pick a template (ez-mcp.py or ez-mcp.ts), and run it directly with uv run ez-mcp.py or deno run --allow-all ez-mcp.ts. Alternatively, copy the template file, rename it, customize the server name and tools, then run your new server. Use MCP Inspector for testing.
Python templates require `uv` (install via `curl -LsSf https://astral.sh/uv/install.sh | sh`). TypeScript templates require Deno (install via `curl -fsSL https://deno.land/install.sh | sh`).
Yes. Each template demonstrates resources (dynamic data sources), tools (LLM‑callable functions), prompts (reusable templates), and environment variable configuration.
Copy the template file, rename it, change the server name in the code, then add your own tools using the documented patterns. Dependencies can be added inline (Python) or imported directly (TypeScript).
Add the command and arguments to your client’s `mcp.json` file. For Python use `"command": "uv", "args": ["run", "/path/to/server.py"]`; for TypeScript use `"command": "deno", "args": ["run", "--allow-all", "/path/to/server.ts"]`.
Yes. The README provides complete examples for a file search tool, a web scraper, and a database query tool in both Python and TypeScript.

Agent-Native Terminal

Automated codebase documentation for GitHub: searchable docs, MCP context for AI agents, doc drift detection, and Friday Cleanup PRs.

Agents make claims. Reelier writes receipts — record an agent's tool-call workflow once, replay it deterministically at 0 tokens, and diff runs to catch drift.

Collide is the multiplayer layer for AI coding agents, and it pays for itself. Expert orchestration across every agent on your repo saves teams tens to hundreds of dollars a day in token burn. Run Claude Code, Codex, Cu

Official Instatus MCP server. Ask what's down and why a monitor is failing, measure uptime for a period, see what went down and for how long, post and resolve incidents, schedule maintenance, put up general notices, see
评论