================== /// MCP /// /// JVM /// ================== [server:online] [protocol:ready]
jvm-mcp-server
by xzq-xu
JVM monitoring MCP server built on Arthas; exposes a Python-based interface for local/remote JVM observability, de-compilation and AI driven performance analysis.
64
12
Open SourceInstallation
1. Prerequisites
• Python 3.10 or higher installed on the host.
• (Optional) Java 17+ if you plan to call JVM-side tooling.
2. Clone the repository
git clone https://github.com/xzq-xu/jvm-mcp-server.git
cd jvm-mcp-server
3. Create and activate a virtual environment (recommended)
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
4. Install dependencies
pip install -r requirements.txt
# or, if a Poetry/PDM file exists
# poetry install
5. (Optional) Configuration
• Export the path of any local models you want to expose:
• Provide an access token if the server is secured:
• Tune server host/port as environment variables or CLI flags.
export MODEL_PATH="/absolute/path/to/model"
export MCP_API_TOKEN="<your-token>"
6. Run the server
# Using the built-in entry point
python -m mcp_server
# or, with Uvicorn (FastAPI)
uvicorn mcp_server.main:app --host 0.0.0.0 --port 8000
7. Verify
curl http://localhost:8000/health
→ {"status":"ok"}
8. Container (optional)
docker build -t jvm-mcp-server .
docker run -p 8000:8000 -e MODEL_PATH=/models jvm-mcp-server
Documentation
License: MIT License
Updated 7/30/2025