jmeter-mcp-server

This is a Model Context Protocol (MCP) server that allows executing JMeter tests through MCP-compatible clients and analyzing test results.

> [!IMPORTANT] > 📢 Looking for an AI Assistant inside JMeter? 🚀 > Check out Feather Wand

• 📊 Execute JMeter tests in non-GUI mode
• 🖥️ Launch JMeter in GUI mode
• 📝 Capture and return execution output
• 📊 Generate JMeter report dashboard

• 📈 Parse and analyze JMeter test results (JTL files)
• 📊 Calculate comprehensive performance metrics
• 🔍 Identify performance bottlenecks automatically
• 💡 Generate actionable insights and recommendations
• 📊 Create visualizations of test results
• 📑 Generate HTML reports with analysis results

1. Install `uv`:

1. Ensure JMeter is installed on your system and accessible via the command line.

⚠️ Important: Make sure JMeter is executable. You can do this by running:

chmod +x /path/to/jmeter/bin/jmeter

1. Install required Python dependencies:

pip install numpy matplotlib

1. Configure the .env file, refer to the .env.example file for details.

# JMeter Configuration
JMETER_HOME=/path/to/apache-jmeter-5.6.3
JMETER_BIN=${JMETER_HOME}/bin/jmeter

# Optional: JMeter Java options
JMETER_JAVA_OPTS="-Xms1g -Xmx2g"

1. Connect to the server using an MCP-compatible client (e.g., Claude Desktop, Cursor, Windsurf)

1. Send a prompt to the server:

Run JMeter test /path/to/test.jmx

1. MCP compatible client will use the available tools:

• 🖥️ execute_jmeter_test: Launches JMeter in GUI mode, but doesn't execute test as per the JMeter design
• 🚀 execute_jmeter_test_non_gui: Execute a JMeter test in