Spring AI ChatMemory Persistence, MCP Server, and SLSA L3: This Week in Spring

Josh Long's latest This Week in Spring is a dense roundup of ecosystem activity, and the through-line is clear: Spring AI is maturing fast, and the tooling around it is catching up.

The most substantive items are three posts from Craig Walls on persisting ChatMemory in Spring AI. One covers JDBC-backed persistence; the other two explore three additional strategies. If you've been treating chat memory as an in-memory afterthought, these recipes show how to make conversational state durable and restart-safe — a prerequisite for any serious production agent.

Also notable: a new Spring Tools MCP server that brings an AI-assisted experience to Spring development in VS Code and Eclipse. MCP (Model Context Protocol) is becoming the standard glue between IDEs and LLMs, and this is a concrete step toward smarter, context-aware coding assistance for Spring developers.

On the supply-chain front, Tanzu (where most of the Spring team sits) now ships SLSA Level 3-compliant dependencies. That's a meaningful security milestone — SLSA L3 means tamper-evident provenance and a hardened build pipeline, something enterprises increasingly demand.

Other highlights include a video on MongoDB-backed repositories in Spring Batch (part of Spring Boot 4.1), a new in-memory vector search engine with out-of-the-box Spring AI integration, and a podcast episode covering Java 27 features.

It's a grab-bag, but the signal is clear: Spring is doubling down on AI integration, and the ecosystem is responding with tools and patterns that make it production-ready.