mcp-llm

Overview

mcp-llm is a JavaScript-based MCP server that enables LLMs to access other LLM capabilities through a standardized interface. It leverages LlamaIndexTS to provide four main tools: generate_code for creating code based on descriptions, generate_code_to_file for directly writing code to files, generate_documentation for automatic documentation generation, and ask_question for querying LLM knowledge. The server offers both manual installation and automated installation via Smithery, making it accessible to different user preferences.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• generate_code

  Generate code based on a description

  Try:Create a function that calculates the factorial of a number in JavaScriptTry:Write a Python function to sort a list of numbers
• generate_code_to_file

  Generate code and write it directly to a file at a specific line number

  Try:Create a function that calculates factorial and save it to factorial.js at line 10Try:Generate a Python sorting function and write it to utils.py
• generate_documentation

  Generate documentation for code

  Try:Generate JSDoc documentation for this JavaScript factorial functionTry:Create documentation for a Python sorting function in Markdown format
• ask_question

  Ask a question to the LLM

  Try:What is the difference between var, let, and const in JavaScript?Try:Explain how to use async functions in Python

Comparable tools

Installation

npx -y @smithery/cli install @sammcj/mcp-llm --client claude

npm install

npm run build

FAQ

What LLM providers does this MCP server support?

The README doesn't specify which LLM providers are supported, but it uses the LlamaIndexTS library which typically supports various LLM providers.

Can I use this with AI clients other than Claude?

The installation guide specifically mentions Claude Desktop, but the underlying MCP protocol should make it compatible with any MCP-compliant client.

Compare mcp-llm with