Tool Calling, MCP & Agent Evaluation
Move from answering questions to taking action — tool calling, hand-rolled agentic patterns, MCP interoperability, and evaluating whether an agent's answers actually hold up.
Tool Calling: Giving the Model the Ability to Act
Move the support bot from answering questions about documents to acting on live data — order lookups, calculations, and REST calls — with Spring AI's @Tool annotation and ToolCallingAdvisor.
Agentic Patterns: ReAct, Plan-and-Execute & Reflection by Hand
Hand-roll multi-step reasoning with ChatClient and ChatMemory — ReAct, Plan-and-Execute, Reflection, sequential chains, and parallel fan-out — and know when to stop hand-rolling.
MCP: Sharing Tools Across Apps, Teams & Models
Expose your Java tools as an MCP server so any team or client can use them, and connect your own agent as an MCP client to tools you didn't write, with Spring AI's annotation-based MCP support.
Agent Evaluation: Catching Regressions Before Customers Do
Stop eyeballing responses. Use Spring AI's Evaluator API — RelevancyEvaluator, FactCheckingEvaluator, and a hand-rolled LLM-as-a-judge — to turn answer quality into a CI gate.