Package Exports

@agentick/google
@agentick/google/dist/index.js

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@agentick/google) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

@agentick/google

Google AI / Vertex AI adapter for Agentick.

Installation

pnpm add @agentick/google

Usage

Factory Pattern (Recommended)

import { google } from "@agentick/google";
import { createApp } from "@agentick/core";

// Google AI Studio
const model = google({
  apiKey: process.env.GOOGLE_API_KEY,
  model: "gemini-2.0-flash",
});

// Use with createApp
const app = createApp(MyAgent, { model });
const session = await app.session();
await session.run({ messages: [{ role: "user", content: [{ type: "text", text: "Hello!" }] }] });

// Or use as JSX component
function MyAgent() {
  return (
    <model temperature={0.7}>
      <System>You are helpful.</System>
      <Timeline />
    </model>
  );
}

// Or call directly
const result = await model.generate({
  messages: [{ role: "user", content: "Hello!" }],
});

Vertex AI

const model = google({
  vertexai: true,
  project: process.env.GCP_PROJECT_ID,
  location: "us-central1",
  model: "gemini-2.0-flash",
  googleAuthOptions: {
    credentials: JSON.parse(process.env.GCP_CREDENTIALS),
  },
});

JSX Component Pattern

import { GoogleModel } from "@agentick/google";

function MyAgent() {
  return (
    <GoogleModel apiKey={process.env.GOOGLE_API_KEY} model="gemini-2.0-flash" temperature={0.7}>
      <System>You are helpful.</System>
      <Timeline />
    </GoogleModel>
  );
}

Configuration

Option	Type	Description
`model`	`string`	Model name (e.g., `gemini-2.0-flash`)
`apiKey`	`string?`	Google AI Studio API key
`vertexai`	`boolean?`	Use Vertex AI instead
`project`	`string?`	GCP project ID (Vertex)
`location`	`string?`	GCP region (Vertex)
`googleAuthOptions`	`object?`	Auth options (Vertex)
`temperature`	`number?`	Sampling temperature
`maxTokens`	`number?`	Maximum tokens to generate

Exports

google(config) - Factory function returning ModelClass
createGoogleModel(config) - Same as google()
GoogleModel - JSX component for declarative usage