Security advisory — Malicious litellm versions 1.82.7 and 1.82.8 were removed from PyPI (potential API key exfiltration). Uninstall them, rotate exposed credentials, and upgrade to a safe release (e.g. 1.82.9+ per upstream). Run pip show litellm to verify. PyPI · README

Build multi-agent systems
without the complexity

A unified, production-ready framework for building intelligent agent applications. From simple automation to complex multi-agent collaboration.

Core capabilities

Everything you need to build production agent systems.

Agent Core

Production-ready execution engine with retry logic and error handling.

|

Orchestration

Graph-based workflow with conditional routing and parallel execution.

Processing...

Memory System

Long-term memory with Mem0 integration, any LLM backend.

Session stored
Context retrieved
Memory consolidated

A2A Communication

Inter-agent messaging protocol

MCP Protocol

Model Context Protocol support

Observability

Tracing, metrics, monitoring

GUI Agent

Desktop automation framework

Tool System

Function decorators & remote tools

Task Validation

Pydantic output parsing

Simple to use

Build complex systems with minimal code.

example.py
from agenticx import Agent, Task, AgentExecutor
from agenticx.llms import OpenAIProvider

agent = Agent(
    id="research-agent",
    name="Research Assistant",
    role="Information gatherer",
    goal="Find and synthesize information"
)

task = Task(
    description="Research latest AI frameworks",
    expected_output="Comprehensive analysis"
)

executor = AgentExecutor(agent=agent, llm=OpenAIProvider())
result = executor.run(task)

Quick start

Get up and running in minutes.

01

Install

$ pip install agenticx
02

Configure

$ export OPENAI_API_KEY="..."
03

Build

$ python your_agent.py

Start building today

Join developers building the next generation of AI applications.