Introduction

• playground  You can see the demo code on the playground.

Set up a fully functional Google Docs Agent powered by OpenAI’s GPT model in no time—using the Agentica CLI.

The Complete Google Docs Agent

Launch the Agentica Setup Wizard with a single command:

npx agentica start google-docs-agent

The wizard guides you through:

• Installing the required packages (e.g., agentica@0.12.14)
• Choosing your package manager and project type
• Selecting the Google Docs controller
• Entering your OPENAI_API_KEY

Once complete, Agentica automatically generates your code, creates a .env file, and installs all dependencies.

Generated Code Overview

The generated Google Docs Agent code looks like this:

import { Agentica } from "@agentica/core";
import typia from "typia";
import dotenv from "dotenv";
import { OpenAI } from "openai";
 
import { GoogleDocsService } from "@wrtnlabs/connector-google-docs";
 
dotenv.config();
 
export const agent = new Agentica({
  vendor: {
    api: new OpenAI({
      apiKey: process.env.OPENAI_API_KEY!,
    }),
    model: "gpt-4o-mini",
  },
  controllers: [
    {
      name: "Google Docs Connector",
      protocol: "class",
      application: typia.llm.application<GoogleDocsService>(),
      execute: new GoogleDocsService(),
    },
  ],
});
 
const main = async () => {
  console.log(await agent.conversate("What can you do?"));
};
 
main();

This code instantly sets up your Google Docs Agent to interact with Google Docs using OpenAI’s GPT model.

Google API Setup

Before running the Google Docs Agent, you need to configure your Google API credentials.

OPENAI_API_KEY=your-openai-api-key
GOOGLE_DOCS_CLIENT_ID=your-google-docs-client-id
GOOGLE_DOCS_CLIENT_SECRET=your-google-docs-client-secret
GOOGLE_DOCS_REFRESH_TOKEN=your-google-docs-refresh-token

To get these credentials:

1. Create a Project in Google Cloud Console:
   • Enable the Google Docs API.
2. Obtain Credentials:
   • Generate OAuth 2.0 credentials to get your Client ID, Client Secret, and Refresh Token.

What This Does

Your Google Docs Agent will:

• Process Google Docs Data: Use the GoogleDocsService connector.
• Leverage OpenAI’s GPT Model: For smart email interactions.
• Maintain Type Safety: With typia.
• Securely Manage Credentials: Using dotenv.

Selecting Specific Functions

You can restrict available functions using TypeScript’s Pick utility. For example, to expose only these functions:

• write
• clear
• readDocs
• update

export const GoogleDocsAgent = new Agentica({
  vendor: {
    api: openai,
    model: "gpt-4o-mini",
  },
  controllers: [
    {
      name: "Google Docs Connector",
      protocol: "class",
      application: typia.llm.application<
        Pick<GoogleDocsService, "write" | "clear" | "readDocs" | "update">
      >(),
      execute: new GoogleDocsService({
        clientId: process.env.GOOGLE_DOCS_CLIENT_ID!,
        clientSecret: process.env.GOOGLE_DOCS_CLIENT_SECRET!,
        secret: process.env.GOOGLE_DOCS_REFRESH_TOKEN!,
      }),
    },
  ],
});

This selective exposure makes your integration even more secure and maintainable.

Available Functions

For a list of available functions in GoogleDocsService, check out the source code:
👉 wrtnlabs/connectors - GoogleDocsService.ts 

Your AI-powered Gmail Agent is now ready for seamless document operation automation! 🚀