Your Agent Needs to Read Before It Writes

by

A few months ago I asked an AI agent to write some Quarkus code for me. The code came back, and it worked. It compiled, it booted, it answered. If the bar were “it works”, the story would end here. Then I read the code: @Produces @ApplicationScoped public PromptGuardAgent promptGuardAgent() { ChatModel guardModel = OllamaChatModel.builder() … Read more

Everyone Says “Context Engineering,” But Thousands of Repositories Say AGENTS.md

by

Eight configuration mechanisms exist for agentic coding tools. A team of researchers went looking for them in 2,853 GitHub repositories, and what came back is one mechanism carrying almost all the weight. The paper is Harness Engineering for Agentic AI Coding Tools: An Exploratory Study. It builds on work published in AIware 2026. The stated … Read more

Quarkus 3.37.0.CR1: extensions start teaching the AI coding agent

by

The release notes for Quarkus 3.37.0.CR1 read like a busy week. A wave of entries adds an “AI skill” to specific extensions, including quarkus-quartz, quarkus-hibernate-reactive, quarkus-redis-cache, and quarkus-security-jpa. One line introduces the ability to get response metadata in a streamed response. One enables Jackson reflection-free serializers by default. Two of these lines, read slowly, change … Read more

LangChain4j Brings the Blackboard Pattern to the JVM

by

The LangChain4j 1.16.0 release notes carry a line that should make anyone running a multi-agent system on the JVM stop scrolling: Introduce Blackboard agentic pattern. Until now, when you wanted agents to coordinate in LangChain4j, you reached for one of the orchestration shapes the library already gave you through AgenticServices: a sequence, a parallel fan-out … Read more