Arthas
Arthas MCP Server is an MCP-based diagnostic toolkit for Java applications, designed for LLM integration. It integrates with Alibaba Arthas so AI assistants can analyze and diagnose Java apps.
What is Arthas?
Arthas is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to arthas mcp server is an mcp-based diagnostic toolkit for java applications, designed for llm integration. it integrates with alibaba arthas so ai assistants can analyze and diagnose java apps.
Arthas MCP Server is an MCP-based diagnostic toolkit for Java applications, designed for LLM integration. It integrates with Alibaba Arthas so AI assistants can analyze and diagnose Java apps.
This server falls under the Developer Tools and Monitoring & Observability categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Arthas MCP Server is an MCP-based diagnostic toolkit for Jav
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx arthasConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Arthas
The Arthas MCP Server is a diagnostic bridge that connects AI assistants to live Java applications through Alibaba's Arthas Java diagnostic tool, exposing nine MCP tools for real-time JVM inspection. It enables AI models to connect to a running Arthas WebConsole, retrieve JVM information, analyze thread states and memory usage, trace method calls, and execute arbitrary Arthas commands — all without requiring the developer to manually run Arthas commands in a terminal. Java developers, DevOps engineers, and SRE teams debugging production performance issues can use this server to let an AI guide the diagnostic process from symptom to root cause.
Prerequisites
- Python 3.10 or later and uv installed
- A running Java application with Alibaba Arthas attached and the WebConsole accessible (default: http://localhost:8563)
- Arthas downloaded and started against the target JVM — see arthas.aliyun.com for Arthas installation
- An MCP client such as Claude Desktop or Cursor
Attach Arthas to your Java application
Download and start Arthas against your running Java process. The WebConsole will be available at http://localhost:8563 by default. This must be running before the MCP server can connect.
curl -O https://arthas.aliyun.com/arthas-boot.jar
java -jar arthas-boot.jarClone the Arthas MCP Server repository
Clone the repository and enter the project directory.
git clone https://github.com/toby1123yjh/arthas_mcp_server
cd arthas_mcp_serverInstall Python dependencies
Use uv to install all required dependencies from the project's lockfile.
uv syncSet the ARTHAS_URL environment variable
Configure the URL where the Arthas WebConsole is listening. The default is http://localhost:8563.
export ARTHAS_URL=http://localhost:8563Configure your MCP client
Add the Arthas MCP server to your Cursor or Claude Desktop configuration. On macOS, the Cursor config is at ~/.cursor/mcp.json.
{
"mcpServers": {
"arthas": {
"command": "uv",
"args": ["--directory", "/path/to/arthas_mcp_server", "run", "python", "main.py"],
"env": {
"ARTHAS_URL": "http://localhost:8563"
}
}
}
}Restart your MCP client and begin diagnosis
Restart Claude Desktop or Cursor. Ask your AI assistant to connect to Arthas and begin diagnosing your Java application.
Arthas Examples
Client configuration
Add this to your MCP client configuration file to run the Arthas MCP server via uv. Replace the directory path with your actual clone location.
{
"mcpServers": {
"arthas": {
"command": "uv",
"args": ["--directory", "/path/to/arthas_mcp_server", "run", "python", "main.py"],
"env": {
"ARTHAS_URL": "http://localhost:8563"
}
}
}
}Prompts to try
Use these prompts to drive AI-assisted Java diagnostics through the nine Arthas MCP tools.
- "Connect to Arthas and show me the current JVM heap memory usage"
- "List all threads that are currently blocked or in WAITING state"
- "Trace method calls to com.example.OrderService.processOrder for the next 10 invocations"
- "Execute the Arthas 'dashboard' command and summarize what you see"
- "Analyze CPU hotspots in the running JVM and identify the most expensive methods"Troubleshooting Arthas
connect_arthas fails with connection refused
Ensure Arthas is running and the WebConsole is accessible. Open http://localhost:8563 in a browser to confirm. If Arthas is running on a different port or host, update ARTHAS_URL accordingly.
uv sync fails or dependencies are not found
Ensure uv is installed (pip install uv or curl -LsSf https://astral.sh/uv/install.sh | sh) and that you are running uv sync from inside the arthas_mcp_server directory where pyproject.toml is located.
Arthas commands return errors about the target class not being found
Verify the fully-qualified class name including the package. Use the execute_arthas_command tool with 'sc com.example.*' to search for loaded classes before running more specific commands.
Frequently Asked Questions about Arthas
What is Arthas?
Arthas is a Model Context Protocol (MCP) server that arthas mcp server is an mcp-based diagnostic toolkit for java applications, designed for llm integration. it integrates with alibaba arthas so ai assistants can analyze and diagnose java apps. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Arthas?
Follow the installation instructions on the Arthas GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Arthas?
Arthas works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Arthas free to use?
Yes, Arthas is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Arthas Alternatives — Similar Developer Tools Servers
Looking for alternatives to Arthas? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Arthas in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Arthas?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.