Stateful agent workflows for Spring AI (graph-based, with retries and recovery)
The spring-agent-flow project introduces a graph-based runtime for building stateful, multi-agent workflows in Java on top of Spring AI, enabling complex AI pipelines with resilience and coordination. It supports retries, circuit breakers, and durable checkpoints to handle real-world challenges like partial failures, rate limits, and human-in-the-loop pauses. The framework simplifies orchestration by eliminating manual routing and loop management, allowing agents to collaborate dynamically within structured workflows.
- ▪spring-agent-flow enables stateful, graph-based agent workflows with retry and recovery mechanisms.
- ▪It supports multi-step pipelines, dynamic routing, and coordination between specialized agents.
- ▪The project includes safety features for message alternation and rate limiting when working with real LLM APIs.
Opening excerpt (first ~120 words) tap to expand
spring-agent-flow Stateful multi-agent orchestration for Spring AI. Design and run long-lived agent workflows with state, retries, and graph execution, all in Java, without manual orchestration code. Independent project — not affiliated with spring-ai-community/agent-client, which is Spring AI Community AgentClient abstraction over CLI agents (Claude Code, Codex, Gemini, etc.). This repository has a different scope: a graph-based runtime for stateful agent workflows on top of Spring AI.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.