Pprof Analyzer MCP Server

Name: Pprof Analyzer MCP Server
Author: ZephyrDeng

@ZephyrDeng

Pprof Analyzer MCP Server について

This is a Model Context Protocol (MCP) server implemented in Go, providing a tool to analyze Go pprof performance profiles.

基本情報

カテゴリ

その他

ライセンス

MIT

ランタイム

トランスポート

stdio

公開者

ZephyrDeng

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "pprof-analyzer-mcp": {
      "command": "docker",
      "args": [
        "build",
        "-t",
        "pprof-analyzer-mcp",
        "."
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Pprof Analyzer MCP Server?

This is a Model Context Protocol (MCP) server implemented in Go that provides tools to analyze Go pprof performance profiles. It integrates with Go's profiling tooling, allowing developers to debug performance issues, memory leaks, goroutine problems, and more.

How to use Pprof Analyzer MCP Server?

Install via go install, build from source, or use the Docker image. Configure the server in your MCP client (e.g., Roo Cline) by adding a JSON entry to .roo/mcp.json that specifies the command to run the executable. The server communicates over the stdio transport.

Key features of Pprof Analyzer MCP Server

Analyze pprof files (CPU, heap, goroutine, allocs, mutex, block)
Output formats: text, markdown, JSON, flamegraph-json
Generate SVG flame graphs via go tool pprof
Compare two profiles to identify regressions/improvements
Detect memory leaks by comparing heap snapshots
Analyze heap time series for memory growth trends

Use cases of Pprof Analyzer MCP Server

Diagnose CPU hotspots in Go applications.
Detect and locate memory leaks using heap snapshots.
Identify goroutine deadlocks or excessive goroutine usage.
Analyze mutex contention and blocking operations.
Compare performance profiles to spot regressions after code changes.

FAQ from Pprof Analyzer MCP Server

What dependencies are required?

Graphviz is required for flame graph generation. The go command must be available for the interactive web UI and flame graph generation.

On which operating systems does the interactive pprof web UI work?

The open_interactive_pprof tool works only on macOS.

How is the server transported?

It uses the stdio transport protocol.

How do I configure the server with my MCP client?

Add a JSON entry to your client’s configuration file (e.g., .roo/mcp.json) specifying the command to run the executable, for example: "command": "pprof-analyzer-mcp".