You are viewing the v1 docs for LangChain, which is currently under active development. Learn more.

​
Overview

LangChain’s createAgent runs on LangGraph’s runtime under the hood. LangGraph exposes a Runtime object with the following information:
  1. Context: static information like user id, db connections, or other dependencies for an agent invocation
  2. Store: a BaseStore instance used for long term memory
  3. Stream writer: an object used for streaming information via the "custom"" stream mode
You can access the runtime information within tools, prompt, and pre and post model hooks.

​
Access

When creating an agent with createAgent, you can specify a contextSchema to define the structure of the context stored in the agent runtime. When invoking the agent, pass the context argument with the relevant configuration for the run: 
import { z } from "zod";
import { createAgent } from "langchain";

const contextSchema = z.object({ 
  userName: z.string(), 
}); 

const agent = createAgent({
  model: "openai:gpt-4o",
  tools: [
    /* ... */
  ],
  contextSchema, 
});

const result = await agent.invoke(
  { messages: [{ role: "user", content: "What's my name?" }] },
  { context: { userName: "John Smith" } } 
);

​
Inside tools

You can access the runtime information inside tools to:
  • Access the context
  • Read or write long term memory
  • Write to the custom stream (ex, tool progress / updates)
Use the runtime parameter to access the Runtime object inside a tool. 
import { z } from "zod";
import { tool } from "langchain";
import { type Runtime } from "@langchain/langgraph";

const contextSchema = z.object({
  userName: z.string(),
});

const fetchUserEmailPreferences = tool(
  async (_, runtime: Runtime<z.infer<typeof contextSchema>>) => {
    const userName = runtime.context?.userName;
    if (!userName) {
      throw new Error("userName is required");
    }

    let preferences = "The user prefers you to write a brief and polite email.";
    if (runtime.store) {
      const memory = await runtime.store?.get(["users"], userName);
      if (memory) {
        preferences = memory.value.preferences;
      }
    }
    return preferences;
  },
  {
    name: "fetch_user_email_preferences",
    description: "Fetch the user's email preferences.",
    schema: z.object({}),
  }
);

​
Inside prompt

Use the runtime parameter to access the Runtime object inside a prompt function. 
import { z } from "zod";
import { createAgent, type AgentState, SystemMessage } from "langchain";
import { type Runtime } from "@langchain/langgraph";

const contextSchema = z.object({
  userName: z.string(),
});

const prompt = (
  state: AgentState,
  runtime: Runtime<z.infer<typeof contextSchema>>
) => {
  const userName = runtime.context?.userName;
  if (!userName) {
    throw new Error("userName is required");
  }

  const systemMsg = `You are a helpful assistant. Address the user as ${userName}.`;
  return [new SystemMessage(systemMsg), ...state.messages];
};

const agent = createAgent({
  model: "openai:gpt-4o",
  tools: [
    /* ... */
  ],
  prompt,
  contextSchema,
});

const result = await agent.invoke(
  { messages: [{ role: "user", content: "What's my name?" }] },
  { context: { userName: "John Smith" } }
);

​
Inside pre and post model hooks

Use the runtime parameter to access the Runtime object inside a pre or post model hook. 
import { z } from "zod";
import { type Runtime } from "@langchain/langgraph";
import { createAgent, type AgentState } from "langchain";

const contextSchema = z.object({
  userName: z.string(),
});

const preModelHook = (
  state: AgentState,
  runtime: Runtime<z.infer<typeof contextSchema>>
) => {
  const userName = runtime.context?.userName;
  if (!userName) {
    throw new Error("userName is required");
  }

  return {
    // ...
  };
};

const postModelHook = (
  state: AgentState,
  runtime: Runtime<z.infer<typeof contextSchema>>
) => {
  const userName = runtime.context?.userName;
  if (!userName) {
    throw new Error("userName is required");
  }

  return {
    // ...
  };
};

const agent = createAgent({
  model: "openai:gpt-4o-mini",
  tools: [
    /* ... */
  ],
  contextSchema,
  preModelHook,
  postModelHook,
});