Installation
Installation guide for Kodey.ai AI Agents.
Getting Started with Kodey.ai
Setting up a Kodey.ai Agent is straightforward, with options to use our CLI tool, SDK, or our developer portal. Kodey.ai requires Node.js version 18 or higher for optimal performance.
Using the Kodey CLI
The Kodey CLI provides the fastest way to create, test, and deploy AI agents.
Step 1: Install the Kodey CLI
Begin by installing the Kodey CLI globally:
npm install -g @kodey/cli
Step 2: Initialize a New Agent
Create a new agent project with the initialization command:
kodey init my-first-agent
Step 3: Configure Agent Settings
Navigate to your new project directory and complete the configuration:
Authentication:
You'll need an API key from your Kodey.ai account. If you don't have one yet, you can create one in the developer portal.
cd my-first-agent
kodey configure
Step 4: Test Your Agent Locally
Run your agent in development mode to test functionality:
kodey dev
Step 5: Deploy Your Agent
When you're ready, deploy your agent to the Kodey.ai platform:
kodey deploy
Using the Kodey.ai SDK
For developers who prefer direct SDK integration, we offer a comprehensive Node.js SDK:
npm install @kodey/sdk
Example Usage:
import { KodeyAgent, Tool } from '@kodey/sdk';
// Initialize the agent
const agent = new KodeyAgent({
apiKey: 'your-api-key',
name: 'My Customer Support Agent',
description: 'Handles customer inquiries and support tickets'
});
// Add capabilities
agent.addTool(
new Tool({
name: 'searchKnowledgeBase',
description: 'Searches the knowledge base for relevant articles',
handler: async (query) => {
// Implementation for knowledge base search
return { articles: ['Article 1', 'Article 2'] };
}
})
);
// Deploy the agent
const deployment = await agent.deploy();
console.log(`Agent deployed: ${deployment.url}`);
Developer Portal Setup
For a visual, no-code experience, you can use our developer portal:
- Sign up or log in at developer.kodey.ai
- Navigate to "Create Agent" in the dashboard
- Configure your agent using the visual builder
- Add tools and integrations from our marketplace
- Test your agent in the sandbox environment
- Deploy with one click when ready
After deployment, you'll receive:
- REST API endpoints
- WebSocket connection details
- HTML snippet for web integration
Deployment Verification
Once deployed, verify your agent is running correctly:
kodey status my-first-agent
Expected Output:
ā
Agent "my-first-agent" is deployed and running
š REST API: https://api.kodey.ai/agents/my-first-agent
š WebSocket: wss://api.kodey.ai/agents/my-first-agent/ws
š» Web Snippet: Available in your dashboard
Performance metrics:
- Average response time: 245ms
- Uptime: 100%
- Usage this month: 1,245 requests
With these steps completed, your AI agent is now running on the Kodey.ai platform and accessible through multiple interfaces.