AI Agent Development Framework
3.3k 2026-04-14

embabel/embabel-agent

A JVM-based framework for building intelligent agents that dynamically blend LLM interactions with custom code and domain models, featuring sophisticated planning capabilities.

Core Features

JVM-native agent development (Kotlin/Java).
Integrates LLM interactions with application code and domain models.
Dynamic, goal-oriented planning with adaptive replanning (OODA loop).
Abstracts agentic concepts like Actions, Goals, and Conditions.
Designed for intelligent path finding towards complex objectives.

Detailed Introduction

Embabel Agent Framework is a robust, JVM-native solution for crafting sophisticated AI agents. It empowers developers to seamlessly combine large language model (LLM) interactions with traditional code and domain-specific models. A key differentiator is its sophisticated, dynamic planning engine, which goes beyond simple state machines by continuously adapting plans based on new information and action outcomes, following an OODA loop approach. This framework simplifies the creation of intelligent, goal-driven applications by abstracting complex agentic concepts, making it ideal for building adaptive and context-aware systems on the JVM.

OSS Alternative

Explore the best open source alternatives to commercial software.

© 2026 OSS Alternative. hotgithub.com - All rights reserved.