It still amazes me how easy & fast it is to create #MCP servers in #Java with #jbang, #langchain4j and a custom skill inside #geminicli

A handful of prompts, and less than 5 minutes later, I had created a working #Wikipedia MCP server!

https://glaforge.dev/posts/2026/04/02/creating-a-wikipedia-mcp-server-in-java-in-a-few-prompts/

AI agents behave much more like terminal users than API clients.

So it’s not surprising that CLI tools often outperform MCP integrations for many workflows.

For Java developers this opens an interesting path:
JBang + small CLI tools as agent capabilities.

I wrote a deep dive here:

https://www.the-main-thread.com/p/ai-agents-cli-tools-jbang-java-architects

#Java #AI #DevTools #JBang #AIagents

Most Java CLI tools stop at mvn compile exec:java.

That works locally. It doesn’t scale.

In this guide I show how to:
- build a fat JAR
- automate GitHub releases
- publish a JBang catalog
- install with one command

No cloning. No Maven required.

Full tutorial:
https://www.the-main-thread.com/p/ship-java-cli-jreleaser-jbang-github-release

#Java #JReleaser #JBang #Maven #OpenSource

RE: https://mastodon.social/@borkdude/116115371711348205

"Cream can run .java source files directly, compiling and caching them automatically. This makes it a fast alternative to #JBang."

#clojure #java #babashka

Java still feels “too heavy” for quick learning? @igfasouza shows how #Java25 + #JEP512, #JBang & notebooks make experimentation fast—then connects it to real hardware via #RaspberryPi + #GraalPy.

Try the stack: https://javapro.io/2026/01/15/java-25-jep-512-jbang-notebooks-graallpy-and-raspberry-pi-for-interactive-learning/

#Agent #skills help 𝗮𝘂𝘁𝗼𝗺𝗮𝘁𝗲 𝗿𝗲𝗽𝗲𝘁𝗶𝘁𝗶𝘃𝗲 𝘁𝗮𝘀𝗸𝘀.

Recently I used #GeminiCLI to easily create #MCP servers in #Java with #jbang & #LangChain4j

I extracted that 𝗸𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 into a 𝗿𝗲𝘂𝘀𝗮𝗯𝗹𝗲 SKILL.md to create more servers!

https://glaforge.dev/posts/2026/02/21/easily-build-a-local-mcp-server-in-java-with-a-skill-in-gemini-cli/

💪 Zero Boilerplate #Java STDIO #MCP Servers with #LangChain4j and #jbang

You just need a single Java file.
JBang installed.
No build! No project structure. Nothing else.

Then configure your custom MCP server to use with your favorite #AI #Agent harness, like #geminicli & friends!

The #langchain4j project recently added a simple #MCP STDIO server.
So you don't need the big guns like
@quarkusio @micronaut & friends for simple MCP servers.

https://glaforge.dev/posts/2026/02/11/zero-boilerplate-java-stdio-mcp-servers-with-langchain4j-and-jbang/

For years we accepted a trade-off: Python for scripts, Java for systems.

With JBang and Java 21, that trade-off is gone.

You can now write single-file Java scripts with near-zero friction and promote them to production without a rewrite.

Hands-on examples: HTTP servers, CLIs, async clients, matrix math.

https://www.the-main-thread.com/p/java-is-the-new-python-jbang-java-21

#Java #JBang #Java21 #DevExperience #SoftwareEngineering #JVM