Introduction

​

The Problem Every AI Developer Faces

You’ve built an incredible AI agent in Python. It uses LangGraph for complex reasoning, leverages powerful tools, and produces amazing results. Your team loves it.

Then reality hits:

Your frontend team needs it in JavaScript, your mobile team wants it in Kotlin, your Unity team wants it in good old C#, your systems team requires it in Rust. The traditional approach? Build separate implementations, REST APIs, WebSocket handlers…

Sound exhausting? That’s because it is.

​

What RunAgent Actually Does

RunAgent fundamentally changes how AI agents work. Here’s the magic:

Your Python function signatures automatically become API contracts for every supported language:

def analyze_market(company: str, timeframe: str = "1year") -> dict:
    return langgraph_agent.invoke({"company": company, "timeframe": timeframe})

The Revolutionary Insight: Change your Python function signature, and all language SDKs automatically adapt. No API versioning, no breaking changes.

​

Getting started

The fastest way to experience the magic is with RunAgent:

1. Install the CLI (includes Python SDK):

   Copy

   pip install runagent
   

2. Create your first agent:

   Copy

   runagent init my_agent
   

3. Deploy locally:

   Copy

   runagent serve my_agent
   

…and you’re ready to access your Python agent from any supported language.

• Head to the quickstart guide to see the full magic
• Browse our framework templates for LangGraph, CrewAI, and more
• Check out core concepts to understand how it works

​

Multi-Language SDK Support

RunAgent provides native-like access to your deployed agents across multiple languages:

​

Framework Support

RunAgent works with any Python-based AI agent framework:

​

Real-Time Streaming Across Languages

When your Python agent streams responses, RunAgent makes it feel native in every target language:

def generate_report(topic: str) -> Iterator[str]:
    yield f"# {topic} Analysis Report\n\n"
    yield "## Executive Summary\n"
    for section in analyze_topic(topic):
        yield f"### {section.title}\n{section.content}\n\n"

def generate_report(topic: str) -> Iterator[str]:
    yield f"# {topic} Analysis Report\n\n"
    yield "## Executive Summary\n"
    for section in analyze_topic(topic):
        yield f"### {section.title}\n{section.content}\n\n"

const stream = await client.runStream({topic: "AI Market Trends"});
for await (const chunk of stream) {
    document.getElementById('report').innerHTML += chunk;
}

let mut stream = client.run_stream(&[("topic", json!("AI Market"))]).await?;
while let Some(chunk) = stream.next().await {
    println!("{}", chunk?);
}

What’s happening: WebSocket connections, real-time data flow, and native iteration patterns—all handled automatically by RunAgent.