Java JAR MCP Server
Enables AI assistants to look up Java class definitions and list dependencies from Maven projects by analyzing local JAR files via the Model Context Protocol.
README
Java JAR MCP Server
An npm package that provides MCP (Model Context Protocol) server functionality for Java Maven projects. This server enables AI assistants to access Java class definitions from Maven dependencies, solving the limitation where Cursor and other VSCode-based IDEs cannot automatically read jar file classes.
Features
- Class Definition Lookup: Find Java class definitions from Maven dependencies
- Dependency Listing: List all Maven dependencies for a project
- JAR Class Search: Search for classes within JAR files using patterns
Installation
Prerequisites
- Node.js 18+
- Maven installed (for Maven projects)
Using via npx
{
"mcpServers": {
"java-jar-mcp": {
"command": "npx",
"args": ["-y", "@biggerball/java-jar-mcp@latest"],
"env": {
"MAVEN_REPO_PATH": "/Users/username/.m2/repository"
}
}
}
}
Usage
After configuration, the MCP server will automatically start and provide the following tools to AI assistants:
Available MCP Tools
find_class_definition
Find the definition of a Java class from Maven dependencies.
Parameters:
className(string, required): Fully qualified class name (e.g.,java.util.ArrayList)pomPath(string, required): Path to pom.xml file (e.g.,"/path/to/project/pom.xml")
Returns: Class definition with source code (if available), methods, and fields
Example:
{
"className": "com.alibaba.fastjson2.filter.BeforeFilter",
"pomPath": "/path/to/project/pom.xml"
}
list_project_dependencies
List all Maven dependencies for a project.
Parameters:
pomPath(string, required): Path to pom.xml file (e.g.,"/path/to/project/pom.xml")
Returns: List of dependencies with JAR file paths
search_classes_in_jar
Search for classes matching a pattern in a JAR file.
Parameters:
jarPath(string): Path to the JAR filesearchPattern(string): Search pattern (supports wildcards*and?)
Returns: List of matching class names
Support & Donation
If you find this project helpful and would like to support its development, I'd be grateful if you could buy me a coffee! ☕
Your support helps me continue improving this project and creating more useful tools for the developer community.

If you enjoy using this project, your support would mean a lot to me. Thank you! 🙏
Local Development Installation
-
Clone the repository
-
Install dependencies and build:
npm install npm run build -
Use local path in MCP configuration:
{ "mcpServers": { "java-jar-mcp": { "command": "node", "args": ["/path/to/java-jar-mcp/dist/index.js"], "env": { "MAVEN_REPO_PATH": "/Users/username/.m2/repository" } } } }
How It Works
The MCP server works through the following steps:
- Parse pom.xml: Read and parse the Maven project's
pom.xmlfile - Locate JAR files: Find corresponding JAR files in the local Maven repository based on Maven coordinates
- Extract class definitions: Extract Java class definition information from JAR files
Development
Project Structure
java-jar-mcp/
├── src/
│ ├── index.ts # MCP server entry point
│ ├── mavenParser.ts # pom.xml parsing
│ ├── jarLocator.ts # JAR file location
│ ├── classExtractor.ts # Class definition extraction
│ ├── tools.ts # MCP tool implementations
│ └── cache.ts # LRU cache implementation
└── dist/ # Compiled MCP server
Building
# Build MCP server
npm run build
# Watch mode build
npm run watch
Testing
npm test
Future Improvements
- Support for Gradle projects
- Enhanced Maven property resolution
- Automatic multi-module project detection and dependency merging
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。