Jons MCP Java Server

Provides Java development capabilities through Eclipse JDT.LS, enabling multi-project workspaces, symbol navigation, and diagnostics.

other

GitHub Stars

other

Language

4 months ago

First Indexed

3 weeks ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "jonmmease-jons-mcp-java": {
      "command": "uv",
      "args": [
        "run",
        "--project",
        "/path/to/jons-mcp-java",
        "jons-mcp-java"
      ],
      "env": {
        "JAVA_HOME": "PATH_TO_JAVA_HOME",
        "JDTLS_HOME": "/path/to/jdtls",
        "JONS_MCP_JAVA_WORKSPACE": "YOUR_WORKSPACE_PATH"
      }
    }
  }
}

You run a Java-focused MCP server that leverages Eclipse JDT.LS to provide development capabilities such as symbol navigation, type information, and diagnostics. It supports multi-project Java workspaces, lazy initialization per project, and safe concurrent access for multiple requests.

How to use

Use this MCP server with an MCP client to enable Java language features in your editor or IDE. You run the server locally or connect to it through your MCP client, then load your Java projects. The server will automatically discover relevant Gradle/Multi-project setups and provide on-demand language services per project.

To start using it from Claude Code or a similar MCP client, add the MCP server using the following command pattern. This registers the server and tells the client how to launch it in stdio mode when needed.

How to install

Prerequisites you must have installed before setting up this MCP server:

Python 3.10 or newer
Java 21 or newer (installation note: on macOS you can install via Homebrew with openjdk@21)
Eclipse JDT.LS (Java language server) installed on your system (example: brew install jdtls on macOS)

Step-by-step installation flow

cd /path/to/jons-mcp-java
uv sync

Additional sections

Configuration and runtime behavior for this MCP server focus on local, on-demand operation and workspace-based discovery. The server starts JDT.LS instances per project as needed and uses an LRU strategy to manage memory when multiple projects are active. It is designed to be safe for concurrent requests.

Available tools

definition

Go to symbol definition

references

Find all references to a symbol

implementation

Find implementations of interfaces or abstract methods

type_definition

Go to type definition

document_symbols

List all symbols in a file

workspace_symbols

Search for symbols across the workspace

diagnostics

Get errors and warnings

hover

Get Javadoc and type information

Built by

VeilStrat

AI signals for GTM teams