Skip to main content

LangChain.js / LangGraph

Connect a LangGraph agent to Leadovaro using @civic/mcp-client with the built-in LangChain adapter. The adapter returns DynamicStructuredTool instances ready to drop into createReactAgent, the legacy AgentExecutor, or any custom LangGraph graph.

Prerequisites​

  • Node.js 20+
  • A Leadovaro account at app.leadovaro.com with a configured toolkit
  • A Leadovaro token and an LLM API key (e.g. Anthropic or OpenAI)

Installation​

pnpm add @civic/mcp-client @langchain/core @langchain/anthropic @langchain/langgraph

Environment Variables​

LEADOVARO_TOKEN=your-leadovaro-token
ANTHROPIC_API_KEY=your-anthropic-key
LEADOVARO_PROFILE_ID=your-profile-id  # optional: lock agent to a specific toolkit
Get Your Credentials

How to generate a Leadovaro token and configure toolkit URL parameters

Connecting to Leadovaro​

Use LeadovaroMcpClient with langchainAdapter() to get tools, then pass them to createReactAgent:

import { LeadovaroMcpClient } from "@civic/mcp-client";
import { langchainAdapter } from "@civic/mcp-client/adapters/langchain";
import { ChatAnthropic } from "@langchain/anthropic";
import { createReactAgent } from "@langchain/langgraph/prebuilt";

const client = new LeadovaroMcpClient({
  auth: { token: process.env.LEADOVARO_TOKEN! },
});

const tools = await client.getTools(langchainAdapter());

const agent = createReactAgent({
  llm: new ChatAnthropic({ model: "claude-sonnet-4-6" }),
  tools,
});

Running the Agent​

async function main() {
  const result = await agent.invoke({
    messages: [{ role: "user", content: "What events do I have today?" }],
  });

  console.log(result.messages.at(-1)?.content);
  await client.close();
}

main().catch(console.error);

Production Configuration​

Lock to a Toolkit​

For production agents, scope to a specific profile:

const client = new LeadovaroMcpClient({
  auth: { token: process.env.LEADOVARO_TOKEN! },
  leadovaroProfile: process.env.LEADOVARO_PROFILE_ID,
});

When a profile is specified, the session is locked by default — the agent cannot switch toolkits or modify its own guardrails. This prevents prompt injection attacks from escaping the defined tool scope.

Multi-Turn Conversations​

Pass a checkpointer to createReactAgent to retain conversation history across invocations:

import { MemorySaver } from "@langchain/langgraph";

const agent = createReactAgent({
  llm: new ChatAnthropic({ model: "claude-sonnet-4-6" }),
  tools,
  checkpointer: new MemorySaver(),
});

const config = { configurable: { thread_id: "session-1" } };

const result = await agent.invoke(
  { messages: [{ role: "user", content: "What events do I have today?" }] },
  config
);

Next Steps​

LangChain (Python)

Python LangGraph integration using leadovaro-mcp-client

Agent Deployment

Production deployment guide: profile locking, URL params, authentication

Guardrails

Constrain what tools your LangGraph agent can use

Audit Trail

Query what your agent did via Leadovaro Chat

Get Credentials

Token generation and URL parameter reference