Basic Agent Example

A simple chat agent that demonstrates the core v1beta builder pattern.

---

Overview

This example shows how to:

• Create an agent with the v1beta builder
• Configure an LLM provider
• Execute a simple query
• Handle responses

---

Complete Code

package main

import (
    "context"
    "fmt"
    "log"
    "github.com/agenticgokit/agenticgokit/v1beta"
)

func main() {
    // Create a basic chat agent
    agent, err := v1beta.NewBuilder("ChatAssistant").
        WithLLM("openai", "gpt-4").
        Build()
    if err != nil {
        log.Fatalf("Failed to create agent: %v", err)
    }

    // Run the agent with a simple query
    result, err := agent.Run(context.Background(), "What is AgenticGoKit?")
    if err != nil {
        log.Fatalf("Agent execution failed: %v", err)
    }

    // Print the response
    fmt.Println("Agent Response:")
    fmt.Println(result.Content)
}

---

Step-by-Step Breakdown

1. Import Dependencies

import (
    "context"
    "fmt"
    "log"
    "github.com/agenticgokit/agenticgokit/v1beta"
)

2. Create Agent with Builder

agent, err := v1beta.NewBuilder("ChatAssistant").
    WithLLM("openai", "gpt-4").
    Build()

Key Points:

• NewBuilder(name) - Creates a new agent builder with a name
• WithLLM(provider, model) - Configures the LLM provider and model
• Build() - Constructs the agent instance

3. Execute Query

result, err := agent.Run(context.Background(), "What is AgenticGoKit?")

Parameters:

• context.Context - For cancellation and timeouts
• query string - The input text for the agent

4. Handle Response

fmt.Println(result.Content)

Result Structure:

• Content - The agent's response text
• Metadata - Additional information about the execution

---

Running the Example

Prerequisites

# Install v1beta
go get github.com/agenticgokit/agenticgokit/v1beta

# Set your API key
export OPENAI_API_KEY="sk-..."

Execute

go run main.go

---

Common Variations

With Custom Configuration

agent, err := v1beta.NewBuilder("CustomAgent").
    WithLLM("openai", "gpt-4").
    WithConfig(&v1beta.Config{
        SystemPrompt: "You are a helpful assistant specialized in Go programming.",
        Temperature:  0.7,
        MaxTokens:    1000,
    }).
    Build()

With Azure OpenAI

agent, err := v1beta.NewBuilder("AzureAgent").
    WithLLM("azure", "gpt-4").
    Build()

Environment Variables:

export AZURE_OPENAI_API_KEY="your-key"
export AZURE_OPENAI_ENDPOINT="https://your-resource.openai.azure.com/"
export AZURE_OPENAI_DEPLOYMENT="gpt-4"

With Ollama (Local)

agent, err := v1beta.NewBuilder("LocalAgent").
    WithLLM("ollama", "llama2").
    Build()

Environment Variables:

export OLLAMA_HOST="http://localhost:11434"

---

Error Handling

Basic Error Handling

agent, err := v1beta.NewBuilder("Agent").
    WithLLM("openai", "gpt-4").
    Build()
if err != nil {
    log.Fatalf("Build failed: %v", err)
}

result, err := agent.Run(ctx, query)
if err != nil {
    // Check error type
    if v1beta.IsLLMError(err) {
        log.Printf("LLM error: %v", err)
    } else if v1beta.IsRetryable(err) {
        log.Printf("Retryable error: %v", err)
    } else {
        log.Fatalf("Fatal error: %v", err)
    }
}

With Context Timeout

ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
defer cancel()

result, err := agent.Run(ctx, "What is Go?")
if err != nil {
    if ctx.Err() == context.DeadlineExceeded {
        log.Println("Request timed out")
    } else {
        log.Printf("Error: %v", err)
    }
}

---

Next Steps

• Streaming Agent - Add real-time streaming responses
• Sequential Workflow - Chain multiple agents
• Memory & RAG - Add memory and knowledge base
• Custom Handlers - Implement custom logic

---

Related Documentation

• Getting Started - Complete beginner guide
• Core Concepts - Understanding agents and builders
• Configuration - All configuration options
• Error Handling - Error patterns and recovery