Template Repository

A ready-to-use example showcasing the Okto Eliza Plugin with complete setup instructions and customization options.

Okto Eliza Sample Agent

View on GitHub

Getting Started

Get started quickly with our pre-configured example that demonstrates the Okto Eliza Plugin in action. This implementation includes authentication setup, basic Web3 operations, and example conversations.

Tip:What's Included

Show

Pre-configured Okto Plugin integration
Google OAuth authentication setup
Example AI agent implementation
Sample Web3 operations and commands
Basic conversation flows

Warning:Essential Setup Required

Show

Before you begin, set up your developer dashboard by making sure you have :

Created your Okto Developer Dashboard account
Get your Okto credits from the dashboard
Obtained your API keys from the dashboard
Enabled the specific Chains and Tokens you plan to use in your application
Optional: Enabled Sponsorship for the desired chains

If you need help, reach out to us on our troubleshooting form and we will contact you.

Prerequisites

Before running the sample application, ensure you have:

Node.js: v23+ (required) - Download
Package Manager: pnpm - Install Guide
Okto API Keys: You need your OKTO_CLIENT_PRIVATE_KEY and OKTO_CLIENT_SWA. Obtain these from the Okto Dashboard.
Google OAuth Credentials: Create OAuth 2.0 credentials in the Google Cloud Console.
LLM Provider: Access to OpenAI, Anthropic, or other supported providers.

Installation

Clone the repository and set up the initial project structure:

bash

# Clone the repository
git clone https://github.com/okto-hq/okto-sdk-eliza-agent-template
cd okto-eliza-ai-agent-example
 
# Install dependencies
pnpm install

Environment Setup

Copy the .env.sample file to .env:

bash

cp .env.sample .env

Edit .env with your credentials:

.env

# Okto specific settings
OKTO_ENVIRONMENT=                # Optional. Defaults to "sandbox". Options: sandbox, staging, production.
OKTO_CLIENT_PRIVATE_KEY=         # Required. Your client private key provided by Okto.
OKTO_CLIENT_SWA=                 # Required. Your client SWA provided by Okto.
 
# Google OAuth settings
GOOGLE_CLIENT_ID=                # Required. Get from https://console.cloud.google.com/
GOOGLE_CLIENT_SECRET=            # Required. Get from https://console.cloud.google.com/
 
# Note: You also need to set the environment variable for the LLM you want to use.

Get your OKTO_CLIENT_PRIVATE_KEY and OKTO_CLIENT_SWA from the Okto Dashboard.

Tip:Setting up Google OAuth

Show

Go to Google Cloud Console
Create OAuth 2.0 credentials for your project
Set the redirect URI to: http://localhost:5000
Save your Client ID and Client Secret in the .env file

Need detailed instructions? Check our Google Console Setup Guide.

Configuration

1. LLM Provider

By default, this sample agent uses Groq as the LLM model provider. To use a different provider:

Open characters/okto.character.json
Update the modelProvider field
Add your provider's API key to .env

characters/okto.character.json

{
  "modelProvider": "groq", // Change to your preferred provider
  // ... other settings
}

Supported providers include OpenAI, Anthropic, and others compatible with Eliza.

2. Agent Character

The agent's personality, behavior, and capabilities are defined in characters/okto.character.json:

characters/okto.character.json

{
  "name": "Okto",
  "plugins": ["okto"],
  "system": "Okto is a helpful assistant that can help you perform blockchain network transactions.",
  "bio": [
    "Okto is a helpful assistant that can help you perform blockchain network transactions."
  ],
  "style": {
    "chat": [
      "be cool, don't act like an assistant",
      "be helpful when asked and be agreeable",
      "be warm and accommodate reasonable requests"
    ]
  },
  "topics": [
    "Blockchain",
    "Blockchain Transactions",
    "Blockchain Network"
  ]
}

You can modify these settings to adjust how your agent:

Responds to queries
Presents information
Handles transactions
Interacts with users

Launch

Start the application and begin testing:

bash

pnpm run start

Try these example commands to test the integration:

"Show me my wallet balance"
"What chains are supported?"
"Help me transfer some tokens"

Next Steps

Now that you have the sample agent running, you can:

Explore our Ideas to Get Started for implementation examples
Check out the Plugin Setup Guide to understand the integration

Need help or have questions? Join our Discord community for support and discussions!

Template Repository

On this page