Crate helios_engine 

Helios Engine

Helios is a powerful and flexible Rust framework for building LLM-powered agents with tool support, chat capabilities, and easy configuration management.

Quick Start

Using as a Library (Direct LLM Calls)

Example

use helios_engine::{LLMClient, ChatMessage};
use helios_engine::config::LLMConfig;

#[tokio::main]
async fn main() -> helios_engine::Result<()> {
    let llm_config = LLMConfig {
        model_name: "gpt-3.5-turbo".to_string(),
        base_url: "https://api.openai.com/v1".to_string(),
        api_key: std::env::var("OPENAI_API_KEY").unwrap(),
        temperature: 0.7,
        max_tokens: 2048,
    };

    let client = LLMClient::new(helios_engine::llm::LLMProviderType::Remote(llm_config)).await?;
    let messages = vec![
        ChatMessage::system("You are a helpful assistant."),
        ChatMessage::user("What is the capital of France?"),
    ];

    let response = client.chat(messages, None, None, None, None).await?;
    println!("Response: {}", response.content);
    Ok(())
}

Overview

use helios_engine::{Agent, Config, CalculatorTool};

#[tokio::main]
async fn main() -> helios_engine::Result<()> {
    let config = Config::from_file("config.toml")?;

    let mut agent = Agent::builder("MyAgent")
        .config(config)
        .system_prompt("You are a helpful assistant.")
        .tool(Box::new(CalculatorTool))
        .build()
        .await?;

    let response = agent.chat("What is 15 * 7?").await?;
    println!("Agent: {}", response);
    Ok(())
}

Features

• Direct LLM Access: Use LLMClient for simple, direct calls to LLM models.
• Agent System: Create intelligent agents with tools and persistent conversation.
• Tool Support: Extensible tool system for adding custom functionality.
• Multi-Provider: Works with OpenAI, Azure OpenAI, local models, and any OpenAI-compatible API.
• Type-Safe: Leverages Rust’s type system for reliability.
• Async: Built on Tokio for high-performance async operations.

Re-exports

pub use agent::Agent;

pub use agent::AgentBuilder;

pub use chat::ChatMessage;

pub use chat::ChatSession;

pub use chat::Role;

pub use config::Config;

pub use config::ConfigBuilder;

pub use config::LLMConfig;

pub use error::HeliosError;

pub use error::Result;

pub use llm::Delta;

pub use llm::LLMClient;

pub use llm::LLMProvider;

pub use llm::LLMRequest;

pub use llm::LLMResponse;

pub use llm::StreamChoice;

pub use llm::StreamChunk;

pub use tools::CalculatorTool;

pub use tools::EchoTool;

pub use tools::FileEditTool;

pub use tools::FileIOTool;

pub use tools::FileListTool;

pub use tools::FileReadTool;

pub use tools::FileSearchTool;

pub use tools::FileWriteTool;

pub use tools::HttpRequestTool;

pub use tools::JsonParserTool;

pub use tools::MemoryDBTool;

pub use tools::QdrantRAGTool;

pub use tools::ShellCommandTool;

pub use tools::SystemInfoTool;

pub use tools::TextProcessorTool;

pub use tools::TimestampTool;

pub use tools::Tool;

pub use tools::ToolParameter;

pub use tools::ToolRegistry;

pub use tools::ToolResult;

pub use tools::WebScraperTool;

pub use tool_builder::ToolBuilder;

pub use rag::Document;

pub use rag::EmbeddingProvider;

pub use rag::InMemoryVectorStore;

pub use rag::OpenAIEmbeddings;

pub use rag::QdrantVectorStore;

pub use rag::RAGSystem;

pub use rag::SearchResult;

pub use rag::VectorStore;

pub use rag_tool::RAGTool;

pub use serve::load_custom_endpoints_config;

pub use serve::start_server;

pub use serve::start_server_with_agent;

pub use serve::start_server_with_agent_and_custom_endpoints;

pub use serve::start_server_with_custom_endpoints;

pub use serve::CustomEndpoint;

pub use serve::CustomEndpointsConfig;

pub use serve::ServerBuilder;

pub use serve::ServerState;

pub use endpoint_builder::delete;

pub use endpoint_builder::get;

pub use endpoint_builder::patch;

pub use endpoint_builder::post;

pub use endpoint_builder::put;

pub use endpoint_builder::EndpointBuilder;

pub use endpoint_builder::EndpointRequest;

pub use endpoint_builder::EndpointResponse;

pub use endpoint_builder::HttpMethod;

pub use forest::AgentId;

pub use forest::CreatePlanTool;

pub use forest::DelegateTaskTool;

pub use forest::ForestBuilder;

pub use forest::ForestMessage;

pub use forest::ForestOfAgents;

pub use forest::SendMessageTool;

pub use forest::ShareContextTool;

pub use forest::SharedContext;

pub use forest::TaskItem;

pub use forest::TaskPlan;

pub use forest::TaskStatus;

pub use forest::UpdateTaskMemoryTool;

pub use auto_forest::AgentConfig;

pub use auto_forest::AutoForest;

pub use auto_forest::AutoForestBuilder;

pub use auto_forest::OrchestrationPlan;

pub use auto_forest::SpawnedAgent;

Modules

agent

Defines the Agent struct and its associated builder, which are central to the Helios Engine.

auto_forest

AutoForest - Automatic orchestration of agent forests for complex tasks.

chat

Provides chat-related functionality, including ChatMessage, ChatSession, and Role.

config

Handles configuration for the engine, including LLM providers and agent settings.

endpoint_builder

Simplified endpoint creation with the builder pattern for custom HTTP endpoints.

error

Defines the custom HeliosError and Result types for error handling.

forest

Forest of Agents - Multi-agent collaboration system.

llm

Manages interactions with Large Language Models (LLMs), including different providers.

rag

RAG (Retrieval-Augmented Generation) system with vector stores and embeddings.

rag_tool

RAG tool implementation for agent use.

serve

Provides HTTP server functionality for exposing OpenAI-compatible API endpoints.

tool_builder

Simplified tool creation with the builder pattern.

tool_macro

Macros for ultra-simple tool creation.

tools

Contains the tool system, including the Tool trait and various tool implementations.

Macros

quick_tool

Quick tool creation with auto-derived types.