AI Agent Development Framework
2.9k 2026-04-26
EvoAgentX/EvoAgentX
EvoAgentX is an open-source framework for building, evaluating, and self-evolving LLM-based AI agents and agentic workflows in an automated and goal-driven manner.
Core Features
Automated Agent Workflow Construction
Integrated Self-Evolution Engine
Built-in Evaluation for Agent Behavior
Plug-and-Play LLM Compatibility
Human-in-the-Loop (HITL) Support
Detailed Introduction
EvoAgentX is an innovative open-source framework designed to streamline the creation, assessment, and optimization of LLM-based AI agents and complex agentic workflows. Moving beyond static prompt engineering, it introduces a self-evolving ecosystem where agents learn and improve through iterative feedback loops. The framework offers automated workflow construction, integrated evaluation, and a powerful self-evolution engine, making it ideal for researchers and developers aiming to deploy robust, adaptive AI solutions with minimal manual intervention and maximum flexibility.