Ross ROSS = Recommend OSS · open-source software intelligence for agents

agents-flex/agents-flex

Agents-flex is A lightweight Java AI agent development framework (positioned as a counterpart to Spring AI). It supports features such as RAG, MCP, Skills, Text2SQL , LLM Wiki, Sub-agents, Web Search, TTS (synchronous and streaming), and STT. observed · 2026-09-03

github.com/agents-flex/agents-flex · homepage · Java · Apache-2.0 (permissive) observed · 2026-09-03

Health v2 · maintenance only

93/100

  • Activity 100
  • Release rhythm 98
  • Longevity 68
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 7.0
  • age_days: 964
  • days_rel: 16
  • days_push: 0
  • n_releases_24m: 33

Full methodology

Adoption not part of the score

1046 stars · 144 forks observed · 2026-09-03

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Agents-Flex is a lightweight, modular Java framework for building AI applications and agents, positioned as a Java counterpart to Spring AI. It provides unified abstractions for chat models, tool calling, durable agent runtimes, RAG pipelines, MCP, Skills with sandboxed execution, multimodal models (image, TTS, STT, video, OCR), Text2SQL, and OpenTelemetry observability.

Use cases

  • build ai agents in java
  • add llm chat to a spring boot app
  • build a rag knowledge base chatbot
  • connect multiple llm providers with one api
  • tool calling and function calling in java
  • streaming chat responses in a web app
  • extract text from pdfs and office docs with ocr
  • run durable long-running ai tasks that survive restarts

When to choose

  • you are a Java/JVM developer building LLM-powered applications or agents
  • you want a Spring AI or LangChain4j alternative that works without Spring
  • you need production features like retries, load balancing, circuit breaking, and observability out of the box
  • you need durable, recoverable agent runs and async task tracking across restarts
  • you want RAG, MCP, Skills, and multimodal capabilities in one modular framework

When to avoid

  • your stack is Python, JavaScript, or another non-JVM language
  • you only need a simple one-off script calling an LLM API
  • you need a ready-made end-user chat application rather than a development framework
  • you are locked to JDK 8 but need the MCP module, which requires JDK 17+

Facets

framework · maturity active

agent-framework llm-inference rag mcp chatbot ocr tts speech-recognition image-processing video-processing web-framework sdk monitoring tracing search-engine workflow-automation prompt-engineering nlp artificial-intelligence large-language-models chatbots machine-learning developer-tools backend apis self-hosted jvm cross-platform self-hosted java spring-ai-alternative tool-calling text2sql skills sub-agents vector-store embedding multimodal openai-compatible streaming async-tasks sandbox opentelemetry ai-agents retrieval-augmented-generation natural-language-processing spring docker

8 sources

Member repositories

RepositoryRoleHealth v2
agents-flex/agents-flexmain93

For agents

markdown · JSON · MCP: product_card(name="agents-flex/agents-flex")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem