Goku

v1.0.0Monitoring & Observabilitystable

Goku is an HTTP load testing application written in Rust

benchmarkbenchmarkinghttpload-testingmcp
Share:
145
Stars
0
Downloads
0
Weekly
0/5

What is Goku?

Goku is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to goku is an http load testing application written in rust

Goku is an HTTP load testing application written in Rust

This server falls under the Monitoring & Observability category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Goku is an HTTP load testing application written in Rust

Use Cases

Goku is an HTTP load testing application written in Rust
jcaromiq

Maintainer

LicenseMIT
Languagerust
Versionv1.0.0
UpdatedMay 13, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx goku

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Goku

Goku is an HTTP load testing tool written in Rust that is also available as an MCP server, enabling AI agents to run performance benchmarks against web APIs and services directly from a conversation. Through its run_benchmark MCP tool, agents can specify a target URL, concurrency level, request count or duration, rate limits, headers, request body, and HTTP/2 support — then receive detailed latency histograms, percentile breakdowns (p50, p95, p99), and throughput metrics without leaving the chat interface. It is well-suited for developers who want to identify API performance bottlenecks or validate service capacity changes through natural language.

Prerequisites

  • Rust toolchain installed (cargo) if building from source, or a supported Linux/macOS/WSL environment for the shell installer
  • Network access to the target HTTP(S) endpoints you want to test
  • An MCP-compatible client such as Claude Desktop or Cursor
  • No API keys required — Goku runs entirely locally
1

Install the Goku MCP server via Cargo

Install the goku-mcp crate from crates.io. This builds the MCP server binary and places it on your PATH.

cargo install goku-mcp
2

Or use the automated shell installer

On Linux, macOS, or WSL you can run the official install script which downloads a pre-built binary for your platform.

curl -sSL https://raw.githubusercontent.com/jcaromiq/goku/v3.0.0/scripts/install_mcp.sh | sh
3

Verify the binary is working

Confirm the goku-mcp binary is installed and accessible before wiring it into your MCP client config.

goku-mcp --version
4

Configure your MCP client

Add the goku-mcp server entry to your claude_desktop_config.json so your AI client can invoke load tests on demand.

{
  "mcpServers": {
    "goku": {
      "command": "goku-mcp",
      "args": []
    }
  }
}
5

Restart your client and run a test load

Restart Claude Desktop and ask the AI to run a quick load test against a safe endpoint to confirm the MCP tool responds correctly.

Goku Examples

Client configuration

claude_desktop_config.json entry for the Goku MCP server after installing goku-mcp via Cargo.

{
  "mcpServers": {
    "goku": {
      "command": "goku-mcp",
      "args": []
    }
  }
}

Prompts to try

Example prompts that trigger the run_benchmark MCP tool to perform HTTP load tests.

- "Run a load test on https://api.example.com/health with 50 concurrent clients for 30 seconds and show me the p95 latency"
- "Benchmark https://api.example.com/search?q=test with 20 clients at a rate of 100 requests per second using HTTP/2"
- "Test https://staging.myapp.com/checkout with 10 concurrent users for 60 seconds and give me a full latency histogram"
- "Run a ramp-up load test on https://api.example.com/orders — start with 5 clients and ramp to 100 over 2 minutes"
- "Compare throughput of https://v1.api.example.com/data vs https://v2.api.example.com/data with identical load"

Troubleshooting Goku

cargo install goku-mcp fails with linker errors

Ensure a C linker is installed. On macOS run xcode-select --install; on Linux install build-essential (Debian/Ubuntu) or base-devel (Arch). Then retry cargo install goku-mcp.

Load test returns connection refused or timeout errors

Verify the target URL is reachable from the machine running goku-mcp. If testing against localhost, ensure the service is running. For HTTPS targets with self-signed certs, the MCP tool accepts an insecure flag to skip TLS verification.

goku-mcp binary not found after Cargo install

Add ~/.cargo/bin to your PATH. Run: export PATH="$HOME/.cargo/bin:$PATH" and add this line to your shell profile (~/.zshrc or ~/.bashrc), then restart your terminal and retry.

Frequently Asked Questions about Goku

What is Goku?

Goku is a Model Context Protocol (MCP) server that goku is an http load testing application written in rust It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Goku?

Follow the installation instructions on the Goku GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Goku?

Goku works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Goku free to use?

Yes, Goku is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Monitoring & Observability MCP Servers

Explore all monitoring & observability servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "goku": { "command": "npx", "args": ["-y", "goku"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Goku?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides