LiteFlow Rule Engine
Lightweight, fast, stable component-based rule engine — orchestrate AI Agents like ordinary components.
- Source repo
- dromara/liteflow
- Stars
- ★ 3.9k
- Last updated
- 28d ago
- License
- Apache-2.0
- Primary language
- Java
- FA score
- 29/100 · Major gaps
At a glance
- How it runs
- Works with
- Universal · cross-platform
- Cost
- Free software; you pay for model usage
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- Java developers needing to decouple complex business logic via rule-driven orchestration.
- Not a fit if
- Teams running JDK 8-17: the AI Agent module needs JDK 21+
- Teams needing English-first docs and community support
- Teams unwilling to learn a DSL and component model
- Source review
- 29/100 · Major gaps 8 safety controls not found
What does this agent do, and when should you use it?
LiteFlow is a modern rules engine framework that combines the best of orchestration and rules engines. It can be used for complex componentized business orchestration. DSL rules drive the whole complex business, enabling smooth refresh hot deployment and supporting embedding of multiple scripting language rules. Starting from v2.16.0, LiteFlow turns an AI Agent into a first-class citizen that can be orchestrated directly into your rules. With over 2000 test cases, it supports JDK 8 to 25, Spring Boot 2.X, 3.X, and the latest 4.X. Rules can be written in XML, JSON, or YML and loaded from various external sources.
LiteFlow parses DSL rule files (XML, JSON, YML) to orchestrate workflows using expressions like THEN, WHEN, IF, etc., enabling synchronous/asynchronous mixed orchestration. The engine executes component invocations, provides context isolation, component retry, and supports script nodes in Groovy, Java, Kotlin, JavaScript, QLExpress, Python, Lua, and Aviator. The AI Agent module (liteflow-react-agent) wraps a full ReAct Agent into a standard component, with multi-turn memory, Skills system, workspace tools, streaming output, and model switching via the model() method. The framework offers elegant hot refresh, so rule changes take effect instantly without restarting the application.
- Java developers needing to decouple complex business logic via rule-driven orchestration.
- Teams requiring dynamic rule updates without restarting in production environments.
- Organizations wanting to integrate AI Agents into existing business rules with orchestration flexibility.
- Developers who want to write rule nodes in multiple scripting languages like Groovy, Python, or JavaScript.
- Enterprise systems needing high performance and stability — LiteFlow is proven in core systems at major companies.
How do you install or deploy this agent?
Add the Maven dependency to your pom.xml:
<dependency>
<groupId>com.yomahub</groupId>
<artifactId>liteflow-spring-boot-starter</artifactId>
<version>2.17.0</version>
</dependency>For the AI Agent module, JDK 21+ is required; add the liteflow-react-agent dependency.
How do you use this agent?
- Define a component by implementing
NodeComponentand overridingprocess(). 2. Write a rule file (e.g., flow.xml) like<chain name="chain1"><then node="a,b,c"/></chain>. 3. Set the rule source in your config:liteflow.rule-source=flow.xml. 4. Execute the chain withFlowExecutor.execute("chain1", context). 5. For AI Agents, declare a component, implement required methods, switch models viamodel(), and orchestrate in rules.
What are this agent's strengths and limitations?
- Simple DSL with a short learning curve; supports complex orchestration patterns.
- Smooth hot refresh of rules without restarting your application.
- Multi-language scripting with seamless Java integration.
- AI Agent is a first-class citizen; multi-model support with one-line model switching.
- AI Agent module requires JDK 21+, which may be a constraint for older environments.
- Documentation and community are primarily in Chinese; English resources are limited.
- Requires learning curve for the DSL and component model, though docs are thorough.
How does this agent compare with similar options?
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| LiteFlow Rule Engine This agent | 29 · Major gaps | Library / SDKFree + model costs | ★ 3.9k | 28d ago | Java | — |
| ReachAI Enterprise Agent Platform | 60 · Some gaps | Self-hosted serviceFree + model costs | ★ 952 | today | Java | OpenAI API |
| Atmosphere Real-Time JVM Agent Engine | 46 · Major gaps | CLIFree + model costs | ★ 3.8k | 1d ago | Java | OpenAI API · Claude API |
| Embabel Agent Framework | 0 · Major gaps | FrameworkFree + model costs | ★ 4.5k | 1d ago | Kotlin | OpenAI API |
How does FollowAgents rate this agent?
Why each dimension lost points
Evidence shows: The repository provides no explicit permission model, user confirmation mechanism, data flow transparency, sensitive data handling, dependency security, external effects, rollback, or source attribution. Therefore all trust criteria score 0.
Evidence shows: README claims over 2000 test cases but provides no test results or execution evidence, so self-consistency scores 1. Dependency availability: README mentions support for multiple storage and scripting languages but provides no dependency list or availability guarantees, scoring 1. Failure messages: No documentation of error handling or failure messages, scoring 0.
Evidence shows: README clearly identifies target audience (Java developers) and scenarios (complex business orchestration), scoring 2. Capability boundaries: README lists features but does not specify limitations or boundaries, scoring 1. Trigger precision: Rule expression examples are clear but trigger conditions are not detailed, scoring 1. Environment fit: README mentions support for Spring Boot 2.X/3.X/4.X and JDK8-25, scoring 2.
Evidence shows: README provides overview, features, scenarios, and AI Agent introduction, good information architecture, scoring 2. Install notes: README does not provide specific installation steps, scoring 1. Naming stability: README mentions version v2.16.0 but no naming conventions, scoring 1. Examples and FAQ: README provides code examples but no FAQ, scoring 2. Known limitations: README mentions AI Agent requires JDK21+ but does not comprehensively list limitations, scoring 1. License: LICENSE file is Apache-2.0, scoring 2. Versioning and changelog: README mentions version but no changelog, scoring 1. Maintenance responsibility: README mentions community-driven but no explicit maintainers, scoring 1.
Evidence shows: Output usability: README describes output features (e.g., streaming) but no specific output format, scoring 1. Marginal value: README emphasizes unique value of AI Agent orchestration, scoring 2. Cost-benefit: README provides no performance benchmarks or cost analysis, scoring 1.
Evidence shows: Claim traceability: Claims in README lack specific evidence or links, scoring 1. Cross-source corroboration: No verification from other sources, scoring 1. Fact-inference separation: Facts and inferences are mixed without clear distinction, scoring 1.
- Not found in source: least-privilege scopingGrant only what the task needs: a dedicated account or read-only token, scoped to specific directories and repos.
- Not found in source: confirmation before actingTurn on (or add) a confirmation step before it acts, and try it in a sandbox or test environment before real data.
- Not found in source: data-flow disclosureWatch which external services it contacts (proxy or firewall logs) and keep sensitive data out until you know where it goes.
- Not found in source: sensitive-data handlingUse dedicated, low-privilege, revocable API keys — never production credentials — and keep secrets out of logs.
- Not found in source: dependency securityPin versions and run a dependency audit (npm audit, pip-audit) before installing; prefer running it in a container.
- Not found in source: disclosed external effectsEstablish which external systems it writes to, sends to or changes, and verify with test accounts or repos before production.
- Not found in source: rollback or recovery pathBack up first, or work on a git branch or snapshot, so its changes can be undone.
- Not found in source: verifiable attributionInstall from the official repo or registry and check the publisher and URL to avoid look-alike packages.
- Static review based only on README and a few config files; source code not examined, so security and functionality cannot be verified.
- Claims in README (e.g., 2000+ test cases) lack verifiable evidence.
- AI Agent module depends on agentscope-java and JDK21+, but no dependency security analysis is provided.
FAQ
How does LiteFlow differ from traditional rule engines like Drools?
How do I integrate different AI models?
model() method in one line, supporting OpenAI, Claude, Gemini, and more, without changing other code.