dromara/liteflow

Lightweight, fast, stable, programmable component-based rule engine — where AI Agents orchestrate just like ordinary components. Uniquely designed DSL: component reuse, sync/async & dynamic orchestration, multi-language scripting, nested rules, hot deployment and smooth refresh. If you can orchestrate LiteFlow, you can orchestrate AI.

What it solves

LiteFlow is a rules engine framework designed to decouple complex, bloated business logic. It allows developers to replace hard-coded, nested conditional logic with a flexible orchestration layer where business processes are defined as a series of components and rules.

How it works

The framework treats every piece of logic as a "component." These components are arranged using a Domain Specific Language (DSL) that supports operators like THEN (sequential), WHEN (parallel), IF (conditional), and SWITCH or FOR (looping/routing). Rules can be written in XML, JSON, or YML and can be hot-refreshed from various sources (like Nacos, Redis, or Zookeeper) without restarting the application.

Who it’s for

It is primarily for Java developers (JDK 8 to 25) using Spring Boot or other Java frameworks who need to manage highly complex business workflows and multi-threaded process choreography.

Highlights

  • AI Agent Integration: Wraps full ReAct (Reasoning + Acting) Agents as standard components, allowing AI to be orchestrated alongside traditional business nodes.
  • Hot Deployment: Rules can be updated instantly without application restarts, even under high concurrency.
  • Multi-Language Support: Supports embedding script rules in Groovy, Java, Kotlin, JavaScript, Python, Lua, and others.
  • Flexible Storage: Rules can be loaded from structured databases, Nacos, Etcd, Zookeeper, Apollo, and Redis.
  • Enterprise Ready: Includes built-in monitoring, component retry mechanisms, and context isolation for high-concurrency data flow.

Related

  • Project
    rulego/rulego

    A lightweight, high-performance embedded rule engine for Go that allows dynamic orchestration of business logic through component-based rule chains.

  • Project
    OpenDCAI/DataFlow

    DataFlow is an open‑source Python framework that lets you build, run, and share low‑code pipelines for generating, cleaning, and evaluating LLM training data. It provides a library of reusable operators, a visual WebUI, an AI‑assistant for auto‑creating pipelines, and a Ray‑based distributed execution layer.

  • Project
    agents-flex/agents-flex

    A lightweight, modular AI application framework for Java that provides unified abstractions for LLMs, RAG, and agents, with production-ready tooling for Spring Boot integration.

  • Project
    langgraph4j/langgraph4j

    A Java library for building stateful, multi-agent AI applications using cyclical graphs, designed to work with LangChain4j and Spring AI.

  • Project
    alikon-art/DeterminFlow

    DeterminFlow is an open‑source, AGPL‑licensed framework for building, running, and monitoring production‑grade AI workflows. It breaks complex pipelines into isolated LLM‑Agent nodes, scripts, API calls and human approvals, providing versioning, checkpoint‑based recovery, per‑node token accounting, and a visual editor. The tool is aimed at content generation, data‑centric AI pipelines, and enterprise automation, and can be installed via desktop binaries, run from source, or deployed with Docker.