litemcp
A TypeScript framework for building MCP servers elegantly
Documentation
LiteMCP
A TypeScript framework for building MCP (Model Context Protocol) servers elegantly
> Deprecate notice:
>
> Since the official SDK already offers a high-level API similar to this project, this project will no longer be maintained.
>
> If you like the CLI commands in this project, you can use them independently: https://github.com/wong2/mcp-cli
Features
- Simple Tool, Resource, Prompt definition
- Full TypeScript support
- Built-in logging
- Built-in error handling
- Built-in CLI for testing and debugging
- Built-in support for SSE transport
Installation
npm install litemcp zodQuickstart
import { LiteMCP } from "litemcp";
import { z } from "zod";
const server = new LiteMCP("demo", "1.0.0");
server.addTool({
name: "add",
description: "Add two numbers",
parameters: z.object({
a: z.number(),
b: z.number(),
}),
execute: async (args) => {
return args.a + args.b;
},
});
server.addResource({
uri: "file:///logs/app.log",
name: "Application Logs",
mimeType: "text/plain",
async load() {
return {
text: "Example log content",
};
},
});
server.start();You can test the server in terminal with:
npx litemcp dev server.jsCore Concepts
Tools
Tools in MCP allow servers to expose executable functions that can be invoked by clients and used by LLMs to perform actions.
server.addTool({
name: "fetch",
description: "Fetch the content of a url",
parameters: z.object({
url: z.string(),
}),
execute: async (args) => {
const content = await fetchWebpageContent(args.url);
return content;
},
});Resources
Resources represent any kind of data that an MCP server wants to make available to clients. This can include:
- File contents
- Screenshots and images
- Log files
- And more
Each resource is identified by a unique URI and can contain either text or binary data.
server.addResource({
uri: "file:///logs/app.log",
name: "Application Logs",
mimeType: "text/plain",
async load() {
return {
text: await readLogFile(),
};
},
});You can also return binary contents in `load`:
async load() {
return {
blob: 'base64-encoded-data'
}
}Prompts
Prompts enable servers to define reusable prompt templates and workflows that clients can easily surface to users and LLMs. They provide a powerful way to standardize and share common LLM interactions.
server.addPrompt({
name: "git-commit",
description: "Generate a Git commit message",
arguments: [
{
name: "changes",
description: "Git diff or description of changes",
required: true,
},
],
load: async (args) => {
return `Generate a concise but descriptive commit message for these changes:\n\n${args.changes}`;
},
});Logging
You can send log messages to the client with `server.logger`
server.addTool({
name: "download",
description: "Download a file from a url",
parameters: z.object({
url: z.string(),
}),
execute: async (args) => {
server.logger.info("Downloading file", { url: args.url });
// ...
server.logger.info("Downloaded file", { url: args.url });
return response;
},
});The `logger` object has the following methods:
- `debug(message: string, context?: JsonValue)`
- `info(message: string, context?: JsonValue)`
- `warn(message: string, context?: JsonValue)`
- `error(message: string, context?: JsonValue)`
Running Your Server
Debugging with `mcp-cli`
The fastest way to test and debug your server is with `mcp-cli`:
npx litemcp dev server.js
npx litemcp dev server.ts // ts files are also supportedThis will run your server with `mcp-cli` for testing and debugging your MCP server in the terminal.
Inspect with `MCP Inspector`
Another way is to use the official `MCP Inspector` to inspect your server with a Web UI:
npx litemcp inspect server.jsSSE Transport
The servers are running with `stdio` transport by default. You can also run the server with SSE mode:
server.start({
transportType: "sse",
sse: {
endpoint: "/sse",
port: 8080,
},
});This will start the server and listen for SSE connections on http://localhost:8080/sse.
You can then connect to the server with SSE transport in the client.
Showcase
> If you've developed a server using LiteMCP, please submit a PR to showcase it here!
- https://github.com/wong2/mcp-jina-reader
- https://github.com/nloui/paperless-mcp
Roadmap
- Add support for Resource Templates
Related
- mcp-cli - A CLI for testing and debugging MCP servers
- mcpservers.org - A curated list of MCP servers
- FastMCP - A Python library for MCP server development, inspiration for this project
Frequently asked questions
What is litemcp?
litemcp is A TypeScript framework for building MCP servers elegantly
How do I install litemcp?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is litemcp open source?
Yes, it is hosted on GitHub at https://github.com/wong2/litemcp and has 179 stars.
Related MCP tools
The go-to web for your AI coding agent — local-first search, fetch, crawl & research over MCP. No API keys, no cloud, $0/query. Public beta.
MCP server that enables AI assistants to interact with Google Gemini CLI, leveraging Gemini's massive token window for large file analysis and codebase understanding
Open-source cross-agent memory layer for coding agents via MCP. Compatible with Claude Code, Codex, Cursor, Windsurf, Gemini CLI, Antigravity, OpenClaw, Hermes Agent, Oh-my-Pi, Pi, Copilot, Kiro, OpenCode, and Trae.
Composio equips your AI agents & LLMs with 100+ high-quality integrations via function calling for the Model Context Protocol. Enhance AI assistants with powerf
📦 Repomix is a powerful tool that packs your entire repository into a single, AI-friendly file. Perfect for when you need to feed your codebase to Large Lan...
The TypeScript AI agent framework. ⚡ Assistants, RAG, observability. Supports any LLM: GPT-4, Claude, Gemini, Llama. Built for the Model Context Protocol to enh
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP