agent_with_rag/
agent_with_rag.rs

1/// Example: Using the Agent with RAG (Retrieval-Augmented Generation)
2///
3/// This example demonstrates:
4/// - Document embedding and storage in Qdrant vector database
5/// - Semantic search with vector similarity
6/// - RAG workflow for context-aware responses
7///
8/// Prerequisites:
9/// 1. Qdrant running locally: docker run -p 6333:6333 qdrant/qdrant
10/// 2. OpenAI API key for embeddings: export OPENAI_API_KEY=your-key
11
12use helios_engine::{Agent, Config, QdrantRAGTool};
13
14#[tokio::main]
15async fn main() -> helios_engine::Result<()> {
16    println!("🚀 Helios Engine - Agent with RAG Example");
17    println!("==========================================\n");
18
19    // Check for required environment variables
20    let embedding_api_key = std::env::var("OPENAI_API_KEY")
21        .unwrap_or_else(|_| {
22            println!("⚠ Warning: OPENAI_API_KEY not set. Using placeholder.");
23            "your-api-key-here".to_string()
24        });
25
26    // Load configuration
27    let config = Config::from_file("config.toml").unwrap_or_else(|_| {
28        println!("⚠ No config.toml found, using default configuration");
29        Config::new_default()
30    });
31
32    // Create RAG tool with Qdrant backend
33    let rag_tool = QdrantRAGTool::new(
34        "http://localhost:6333",           // Qdrant URL
35        "helios_knowledge",                 // Collection name
36        "https://api.openai.com/v1/embeddings", // Embedding API
37        embedding_api_key,                  // API key
38    );
39
40    // Create agent with RAG tool
41    let mut agent = Agent::builder("KnowledgeAgent")
42        .config(config)
43        .system_prompt(
44            "You are a helpful assistant with access to a RAG (Retrieval-Augmented Generation) system. \
45             You can store documents and retrieve relevant information to answer questions. \
46             When answering questions, first search for relevant documents, then provide informed answers based on the retrieved context."
47        )
48        .tool(Box::new(rag_tool))
49        .max_iterations(10)
50        .build()
51        .await?;
52
53    println!("✓ Agent created with RAG capabilities\n");
54
55    // Example 1: Add knowledge to the database
56    println!("Example 1: Adding Documents to Knowledge Base");
57    println!("==============================================\n");
58
59    let response = agent
60        .chat(
61            "Store this information: Rust is a systems programming language that runs blazingly fast, \
62             prevents segfaults, and guarantees thread safety. It was created by Mozilla Research."
63        )
64        .await?;
65    println!("Agent: {}\n", response);
66
67    let response = agent
68        .chat(
69            "Store this: Python is a high-level, interpreted programming language known for its \
70             clear syntax and readability. It was created by Guido van Rossum in 1991."
71        )
72        .await?;
73    println!("Agent: {}\n", response);
74
75    let response = agent
76        .chat(
77            "Store this: JavaScript is a programming language commonly used for web development. \
78             It allows developers to create interactive web pages and runs in web browsers."
79        )
80        .await?;
81    println!("Agent: {}\n", response);
82
83    // Example 2: Semantic search - ask questions
84    println!("\nExample 2: Semantic Search and Q&A");
85    println!("===================================\n");
86
87    let response = agent
88        .chat("What programming language is known for preventing segfaults?")
89        .await?;
90    println!("Agent: {}\n", response);
91
92    let response = agent
93        .chat("Tell me about the programming language created in 1991")
94        .await?;
95    println!("Agent: {}\n", response);
96
97    // Example 3: Multi-document retrieval
98    println!("\nExample 3: Multi-Document Retrieval");
99    println!("====================================\n");
100
101    let response = agent
102        .chat("Search for information about programming languages and summarize what you find")
103        .await?;
104    println!("Agent: {}\n", response);
105
106    // Example 4: Adding documents with metadata
107    println!("\nExample 4: Documents with Metadata");
108    println!("===================================\n");
109
110    let response = agent
111        .chat(
112            "Store this with metadata: \
113             The Helios Engine is a Rust framework for building LLM agents. \
114             Metadata: category=framework, language=rust, year=2024"
115        )
116        .await?;
117    println!("Agent: {}\n", response);
118
119    println!("\n✅ Example completed successfully!");
120    println!("\n💡 Key Features Demonstrated:");
121    println!("  • Document embedding with OpenAI embeddings");
122    println!("  • Vector storage in Qdrant database");
123    println!("  • Semantic search with cosine similarity");
124    println!("  • RAG workflow for context-aware answers");
125    println!("  • Metadata support for document organization");
126    
127    println!("\n📝 RAG Use Cases:");
128    println!("  • Question answering over custom knowledge bases");
129    println!("  • Document search and retrieval");
130    println!("  • Building chatbots with domain-specific knowledge");
131    println!("  • Information extraction from large document sets");
132    
133    println!("\n🔧 Setup Instructions:");
134    println!("  1. Start Qdrant: docker run -p 6333:6333 qdrant/qdrant");
135    println!("  2. Set API key: export OPENAI_API_KEY=your-key");
136    println!("  3. Run example: cargo run --example agent_with_rag");
137
138    Ok(())
139}
agent_with_rag/agent_with_rag.rs

agent_with_rag/
agent_with_rag.rs
