Mcp Agent Kit

@dominiquekossi

About Mcp Agent Kit

a complete and intuitive SDK for building MCP Servers, MCP Agents, and LLM integrations (OpenAI, Claude, Gemini) with minimal effort.

#mcp-agent-kit #sdk #ai-agents

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "mcp-agent-kit": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-agent-kit"
      ],
      "env": {
        "MCP_SERVER_NAME": "my-server",
        "MCP_PORT": "7777",
        "LOG_LEVEL": "info",
        "OPENAI_API_KEY": "sk-...",
        "ANTHROPIC_API_KEY": "sk-ant-...",
        "GEMINI_API_KEY": "...",
        "OLLAMA_HOST": "http://localhost:11434"
      }
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Mcp Agent Kit?

Mcp Agent Kit is a TypeScript package that simplifies creating MCP (Model Context Protocol) servers, AI agents, intelligent LLM routers, chatbots with conversation memory, and HTTP API helpers with built-in retry and timeout. It supports multiple LLM providers and is designed for developers who want to quickly build production-ready agents and tools.

How to use Mcp Agent Kit?

Install the package via npm (npm install mcp-agent-kit), then use functions like createAgent, createMCPServer, createChatbot, and createLLMRouter to build components. Configuration is optional via environment variables or code parameters. Examples are available in the /examples directory and can be run with npx ts-node.

Key features of Mcp Agent Kit

Zero config with smart defaults
Multi-LLM provider support (OpenAI, Anthropic, Gemini, Ollama)
Full TypeScript support and autocomplete
Built-in retry, timeout, and error handling
One-line setup for complex features (agents, servers, chatbots)
Extensible with custom providers and middleware

Use cases of Mcp Agent Kit

Create an MCP server to expose tools and resources to clients
Build an AI agent with function calling and system prompts
Develop a chatbot with automatic conversation memory and history limits
Route requests to different LLMs based on input length or content rules
Simplify HTTP requests with automatic retry and configurable timeout

FAQ from Mcp Agent Kit

Which LLM providers are supported?

OpenAI, Anthropic, Gemini, and Ollama are supported out of the box.

Do I need API keys for all providers?

API keys are required for OpenAI, Anthropic, and Gemini. Ollama runs locally and does not need an API key.

Can I use WebSocket transport for MCP servers?

Yes – call server.start("websocket") to start the server on WebSocket instead of stdio.

How do I configure environment variables?

Set variables like OPENAI_API_KEY or LOG_LEVEL in a .env file or in your environment. The package automatically loads .env files using dotenv.

What is the default transport for MCP servers?

The default transport is stdio, unless you specify "websocket" as an argument to start().

Frequently asked questions

Which LLM providers are supported?

OpenAI, Anthropic, Gemini, and Ollama are supported out of the box.

Do I need API keys for all providers?

API keys are required for OpenAI, Anthropic, and Gemini. Ollama runs locally and does not need an API key.

Can I use WebSocket transport for MCP servers?

Yes – call `server.start("websocket")` to start the server on WebSocket instead of stdio.

How do I configure environment variables?

Set variables like `OPENAI_API_KEY` or `LOG_LEVEL` in a `.env` file or in your environment. The package automatically loads `.env` files using dotenv.

What is the default transport for MCP servers?

The default transport is stdio, unless you specify `"websocket"` as an argument to `start()`.

Comments

Basic information

More AI & Agents MCP servers

Perplexity MCP Server

DaInfernalCoder

A Model Context Protocol (MCP) server for research and documentation assistance using Perplexity AI. Won 1st @ Cline Hackathon

295

Shell and Coding agent for Claude and other mcp clients

rusiaaman

Shell and coding agent on mcp clients

666

Model Context Protocol for Unreal Engine

chongdashu

Enable AI assistant clients like Cursor, Windsurf and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP).

2.0K

1Panel

1Panel-dev

🔥 1Panel is a modern, open-source VPS control panel — and the only one with native AI agent support. Run Ollama models, deploy OpenClaw agents, and manage your entire server stack from one clean web interface.

36.1K

MCP-NixOS - Because Your AI Assistant Shouldn't Hallucinate About Packages

utensils

MCP-NixOS - Model Context Protocol Server for NixOS resources

739