This example shows how to load and use an agent with a JSON toolkit.
npm
npm install @langchain/openai @langchain/core
import * as fs from "fs";
import * as yaml from "js-yaml";
import { OpenAI } from "@langchain/openai";
import { JsonSpec, JsonObject } from "langchain/tools";
import { JsonToolkit, createJsonAgent } from "langchain/agents";

export const run = async () => {
  let data: JsonObject;
  try {
    const yamlFile = fs.readFileSync("openai_openapi.yaml", "utf8");
    data = yaml.load(yamlFile) as JsonObject;
    if (!data) {
      throw new Error("Failed to load OpenAPI spec");
    }
  } catch (e) {
    console.error(e);
    return;
  }

  const toolkit = new JsonToolkit(new JsonSpec(data));
  const model = new OpenAI({ temperature: 0 });
  const executor = createJsonAgent(model, toolkit);

  const input = `What are the required parameters in the request body to the /completions endpoint?`;

  console.log(`Executing with input "${input}"...`);

  const result = await executor.invoke({ input });

  console.log(`Got output ${result.output}`);

  console.log(
    `Got intermediate steps ${JSON.stringify(
      result.intermediateSteps,
      null,
      2
    )}`
  );
};