Model Context Protocol (MCP) 1.15.2+

A standardized boundary for discovering and executing tools, resources, and prompt templates across any compliant server.

Table of contents

1. Unified Execution Pattern
2. Infrastructure & Monitoring
   1. Multi-Server Orchestration
3. API Reference
   1. MCP (The Host)
   2. MCPResource
   3. MCPPrompt

---

NodeLLM acts as an MCP Host, providing a standardized boundary for discovering and executing tools, resources, and prompt templates across any compliant server. By implementing the MCP specification, NodeLLM decouples your agent’s business logic from the specific API nuances of external services.

npm install @node-llm/mcp

---

Unified Execution Pattern

The unique advantage of MCP in NodeLLM is the ability to orchestrate tools, resources, and prompts in a single cohesive workflow using the fluent Chat API.

import { MCP } from "@node-llm/mcp";
import { NodeLLM } from "@node-llm/core";

// 1. Initialize the connection
const github = await MCP.connect({
  command: "npx",
  args: ["-y", "@modelcontextprotocol/server-github"]
});

// 2. Discover all capabilities (with optional namespacing)
const { tools, resources, prompts } = await github.discover({ prefix: "gh_" });

// 3. Resolve context and expert intent
const reviewPrompt = prompts.find(p => p.name.includes("review"));
const sourceFile = resources.find(r => r.name.includes("MCP.ts"));

// 4. Orchestrate in a single session
const { messages } = await reviewPrompt.get({ 
  code: await sourceFile.readText() 
});

const response = await NodeLLM.chat("gpt-4o")
  .withTools(tools)
  .addMessages(messages)
  .ask("Analyze the code and execute tools to fix any bugs found.");

---

Infrastructure & Monitoring

The MCP class provides a clean, event-driven interface for observing server activity.

const mcp = (await MCP.connect(config))
  .onLog(({ level, message }) => console.log(`[${level}] ${message}`))
  .onProgress(({ progress, total }) => console.log(`Progress: ${progress}/${total}`))
  .onError(err => console.error("Protocol Error:", err));

Multi-Server Orchestration

You can combine tools from multiple servers effortlessly. NodeLLM treats them as pluggable Tool Sources.

const mcps = await MCP.connectAll({
  docs: { command: "npx", args: ["-y", "@modelcontextprotocol/server-filesystem", "./docs"] },
  search: { command: "npx", args: ["-y", "@modelcontextprotocol/server-brave-search"] }
});

const chat = llm.chat().withTools([
  ...(await mcps.docs.discoverTools()),
  ...(await mcps.search.discoverTools())
]);

---

API Reference

MCP (The Host)

Method	Description
static connect(config)	Connects to a local server process (Stdio). Accepts command, args, and env.
static connectSSE(config)	Connects to a remote server (HTTP/SSE). Accepts url.
discover(options?)	Master discovery method. Returns tools, resources, and prompts.
onLog(handler)	Registers a listener for server logs.
onProgress(handler)	Registers a listener for progress notifications.
close()	Gracefully disconnects from the server.

MCPResource

Method	Description
read()	Returns the raw ResourceContent from the server.
readText()	Helper to fetch and concatenate all text parts of a resource.

MCPPrompt

Method	Description
get(args?)	Resolves the prompt template with arguments and returns Message[].

---