MCP Tools

An open protocol that enables seamless integration between LLM applications and external data sources and tools.

What is Model Context Protocol (MCP)? The Model Context Protocol (MCP) is an open protocol that allows seamless integration between LLM applications and external data sources and tools.

While using BeaconLabs, you can directly use tools developed with MCP (MCP Servers). This enables you to create various agents and handle tasks using both official tools developed by different companies and community-developed tools, providing access to an extensive tool ecosystem.

Server Lists There are several MCP server lists that allow you to view various MCP servers and learn how to use them. Some of these are:

• https://glama.ai/mcp/servers

• https://www.mcp.run/

• https://smithery.ai/

Server Types

• NPX Based: Requires the installation of Node.js to use these tools.

• UVX: Python-based tools that do not require any installation.

• Docker: Tools that can be easily run with Docker if you have it installed, providing great isolation.

Connecting an Example MCP Server BeaconLabs uses a structure involving class creation for connecting to MCP servers. This structure allows you to create multiple integrations and use them whenever needed.

Here’s an example of integrating the HackerNews MCP:

pythonCopyEdit# Hackernews MCP Server
class HackerNewsMCP:
    command = "uvx"
    args = ["mcp-hn"]

task = Task(
  "Get the latest technical news",
  tools=[HackerNewsMCP]
)

And here’s the GitHub MCP integration:

pythonCopyEdit# GitHub MCP Server - Requires Node.JS
class GitHubMCP:
    command = "npx"
    args = ["-y", "@modelcontextprotocol/server-github"]
    env = {
      "GITHUB_PERSONAL_ACCESS_TOKEN": "<YOUR_TOKEN>"
    }

task = Task(
  "Check the repositories",
  tools=[GitHubMCP]
)