Nutrient Document Engine MCP Server

import { AIMessage } from "@langchain/core/messages"; import { RunnableConfig } from "@langchain/core/runnables"; import { MessagesAnnotation, StateGraph } from "@langchain/langgraph"; import { ToolNode } from "@langchain/langgraph/prebuilt"; import { ConfigurationSchema, ensureConfiguration } from "./configuration.js"; import { MultiServerMCPClient } from "@langchain/mcp-adapters"; import { ChatOpenAI } from "@langchain/openai"; const MCP_SERVER_URL = process.env.MCP_SERVER_URL || "http://localhost:5100/mcp"; const mcpClient = new MultiServerMCPClient({ // Global tool configuration options throwOnLoadError: true, prefixToolNameWithServerName: false, useStandardContentBlocks: true, // Server configuration mcpServers: { // Document engine server documentEngine: { transport: "http", url: MCP_SERVER_URL, }, }, }); // Define the function that calls the model async function callModel( state: typeof MessagesAnnotation.State, config: RunnableConfig ): Promise<typeof MessagesAnnotation.Update> { /** Call the LLM powering our agent. **/ const configuration = ensureConfiguration(config); const tools = await mcpClient.getTools(); console.log(tools); // Feel free to customize the prompt, model, and other logic! const model = new ChatOpenAI({ model: "gpt-4.1-mini", temperature: 0.1, }).bindTools(tools); const response = await model.invoke([ { role: "system", content: configuration.systemPromptTemplate.replace( "{system_time}", new Date().toISOString() ), }, ...state.messages, ]); // We return a list, because this will get added to the existing list return { messages: [response] }; } // Define the function that determines whether to continue or not function routeModelOutput(state: typeof MessagesAnnotation.State): string { const messages = state.messages; const lastMessage = messages[messages.length - 1]; // If the LLM is invoking tools, route there. if ((lastMessage as AIMessage)?.tool_calls?.length || 0 > 0) { return "tools"; } // Otherwise end the graph. else { return "__end__"; } } // Define a new graph. We use the prebuilt MessagesAnnotation to define state: // https://langchain-ai.github.io/langgraphjs/concepts/low_level/#messagesannotation const workflow = new StateGraph(MessagesAnnotation, ConfigurationSchema) // Define the two nodes we will cycle between .addNode("callModel", callModel) .addNode("tools", new ToolNode(await mcpClient.getTools())) // Set the entrypoint as `callModel` // This means that this node is the first one called .addEdge("__start__", "callModel") .addConditionalEdges( // First, we define the edges' source node. We use `callModel`. // This means these are the edges taken after the `callModel` node is called. "callModel", // Next, we pass in the function that will determine the sink node(s), which // will be called after the source node is called. routeModelOutput ) // This means that after `tools` is called, `callModel` node is called next. .addEdge("tools", "callModel"); // Finally, we compile it! // This compiles it into a graph you can invoke and deploy. export const graph = workflow.compile({ interruptBefore: [], // if you want to update the state before calling the tools interruptAfter: [], });