Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/mofa-org/mofa/llms.txt

Use this file to discover all available pages before exploring further.

Learn how to build your first AI agent using the MoFA framework. This tutorial will guide you through creating a simple LLM-powered agent that can answer questions about Rust.

​
Prerequisites

Before you begin, ensure you have:
  • Rust stable toolchain (edition 2024 - requires Rust ≥ 1.85)
  • Git for cloning the repository
  • An OpenAI API key or access to a compatible LLM provider
1

Install Rust

Install the Rust toolchain using rustup:
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
rustup default stable
Verify your installation:
rustc --version   # Should be 1.85.0 or newer
cargo --version
2

Create a New Project

Create a new Rust project for your agent:
cargo new my-first-agent
cd my-first-agent
Add MoFA SDK to your Cargo.toml:
[dependencies]
mofa-sdk = { git = "https://github.com/mofa-org/mofa.git" }
tokio = { version = "1", features = ["full"] }
dotenvy = "0.15"
3

Set Up Environment Variables

Create a .env file in your project root to store your API credentials:
OPENAI_API_KEY=sk-...
OPENAI_MODEL=gpt-4o           # optional, default: gpt-4o
The .env file will be automatically loaded by the dotenvy crate.
4

Define Your Agent Structure

Create your agent by implementing the MoFAAgent trait. Here’s a complete example:
src/main.rs
use std::sync::Arc;
use dotenvy::dotenv;
use mofa_sdk::kernel::agent::prelude::*;
use mofa_sdk::llm::{LLMClient, openai_from_env};

struct LLMAgent {
    id: String,
    name: String,
    capabilities: AgentCapabilities,
    state: AgentState,
    client: LLMClient,
}

impl LLMAgent {
    fn new(client: LLMClient) -> Self {
        Self {
            id: "llm-agent-1".to_string(),
            name: "LLM Agent".to_string(),
            capabilities: AgentCapabilities::builder()
                .tag("llm").tag("qa")
                .input_type(InputType::Text)
                .output_type(OutputType::Text)
                .build(),
            state: AgentState::Created,
            client,
        }
    }
}

​
Understanding the Structure

  • id: Unique identifier for your agent
  • name: Human-readable name
  • capabilities: Describes what your agent can do (used for discovery and routing)
  • state: Current lifecycle state of the agent
  • client: LLM client for making API calls
5

Implement the MoFAAgent Trait

Now implement the required trait methods:
src/main.rs
#[async_trait]
impl MoFAAgent for LLMAgent {
    fn id(&self) -> &str {
        &self.id
    }

    fn name(&self) -> &str {
        &self.name
    }

    fn capabilities(&self) -> &AgentCapabilities {
        &self.capabilities
    }

    fn state(&self) -> AgentState {
        self.state.clone()
    }

    async fn initialize(&mut self, _ctx: &AgentContext) -> AgentResult<()> {
        self.state = AgentState::Ready;
        Ok(())
    }

    async fn execute(&mut self, input: AgentInput, _ctx: &AgentContext) -> AgentResult<AgentOutput> {
        self.state = AgentState::Executing;
        
        let answer = self.client
            .ask_with_system("You are a helpful Rust expert.", &input.to_text())
            .await
            .map_err(|e| AgentError::ExecutionFailed(e.to_string()))?;
        
        self.state = AgentState::Ready;
        Ok(AgentOutput::text(answer))
    }

    async fn shutdown(&mut self) -> AgentResult<()> {
        self.state = AgentState::Shutdown;
        Ok(())
    }
}

​
Core Methods Explained

  • initialize: Prepare the agent for execution (load resources, establish connections)
  • execute: The main task execution method - processes input and returns output
  • shutdown: Clean up resources gracefully
6

Create the Main Function

Wire everything together in your main function:
src/main.rs
#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    dotenv().ok();   // Load .env file

    // Create LLM provider and client
    let provider = openai_from_env()?;
    let client   = LLMClient::new(Arc::new(provider));

    // Create and initialize agent
    let mut agent = LLMAgent::new(client);
    let ctx       = AgentContext::new("exec-001");

    agent.initialize(&ctx).await?;

    // Execute a query
    let output = agent.execute(
        AgentInput::text("What is the borrow checker in Rust?"),
        &ctx,
    ).await?;

    println!("{}", output.as_text().unwrap_or("(no answer)"));
    
    // Clean shutdown
    agent.shutdown().await?;
    Ok(())
}
7

Run Your Agent

Build and run your first agent:
cargo run
You should see a response from the LLM explaining the Rust borrow checker!

​
Complete Example

Here’s the complete code for reference:
src/main.rs
use std::sync::Arc;
use dotenvy::dotenv;
use mofa_sdk::kernel::agent::prelude::*;
use mofa_sdk::llm::{LLMClient, openai_from_env};

struct LLMAgent {
    id: String,
    name: String,
    capabilities: AgentCapabilities,
    state: AgentState,
    client: LLMClient,
}

impl LLMAgent {
    fn new(client: LLMClient) -> Self {
        Self {
            id: "llm-agent-1".to_string(),
            name: "LLM Agent".to_string(),
            capabilities: AgentCapabilities::builder()
                .tag("llm").tag("qa")
                .input_type(InputType::Text)
                .output_type(OutputType::Text)
                .build(),
            state: AgentState::Created,
            client,
        }
    }
}

#[async_trait]
impl MoFAAgent for LLMAgent {
    fn id(&self) -> &str { &self.id }
    fn name(&self) -> &str { &self.name }
    fn capabilities(&self) -> &AgentCapabilities { &self.capabilities }
    fn state(&self) -> AgentState { self.state.clone() }

    async fn initialize(&mut self, _ctx: &AgentContext) -> AgentResult<()> {
        self.state = AgentState::Ready;
        Ok(())
    }

    async fn execute(&mut self, input: AgentInput, _ctx: &AgentContext) -> AgentResult<AgentOutput> {
        self.state = AgentState::Executing;
        let answer = self.client
            .ask_with_system("You are a helpful Rust expert.", &input.to_text())
            .await
            .map_err(|e| AgentError::ExecutionFailed(e.to_string()))?;
        self.state = AgentState::Ready;
        Ok(AgentOutput::text(answer))
    }

    async fn shutdown(&mut self) -> AgentResult<()> {
        self.state = AgentState::Shutdown;
        Ok(())
    }
}

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    dotenv().ok();

    let provider = openai_from_env()?;
    let client   = LLMClient::new(Arc::new(provider));

    let mut agent = LLMAgent::new(client);
    let ctx       = AgentContext::new("exec-001");

    agent.initialize(&ctx).await?;

    let output = agent.execute(
        AgentInput::text("What is the borrow checker in Rust?"),
        &ctx,
    ).await?;

    println!("{}", output.as_text().unwrap_or("(no answer)"));
    agent.shutdown().await?;
    Ok(())
}

​
Next Steps

LLM Integration

Learn how to integrate different LLM providers (OpenAI, Anthropic, Ollama, Gemini)

Agent Lifecycle

Understand the full agent lifecycle (pause, resume, interrupt)

Capabilities

Master agent capabilities and state management

API Reference

Explore the complete MoFA API documentation

​
Common Issues

Make sure you’ve created a .env file in your project root with your API key:
OPENAI_API_KEY=sk-...
MoFA requires Rust 1.85 or newer. Update your toolchain:
rustup update stable
If you’re using a proxy or firewall, you may need to configure your network settings. You can also increase the timeout in the LLM configuration.

Build docs developers (and LLMs) love

Get started for freeTalk to us