Skip to content

Getting Started with Code Scalpel

Welcome to Code Scalpel! This section will help you get up and running quickly, whether you're using Claude Desktop, VS Code with Copilot, Cursor, or running Code Scalpel in Docker.

Choose Your Path

  • New to Coding? Start Here!

    Vibe coder who uses AI to code for you? This friendly guide is written just for you!

    Vibe Coder's Guide

  • Quick Start (5 Minutes)

    Already familiar with MCP servers? Jump straight to the fastest setup path.

    Quick start

  • Installation Guides

    Step-by-step installation for your specific IDE or environment.

    Installation

  • Your First Analysis

    Run your first code analysis and understand the results.

    First analysis

  • Understanding Tiers

    Learn about Community, Pro, and Enterprise features.

    Tiers

What You'll Need

Before you begin, make sure you have:

  • Python 3.10+ installed on your system
  • An MCP-compatible AI assistant (Claude Desktop, VS Code with Copilot, Cursor, or similar)
  • A code project to analyze (any Python, JavaScript, TypeScript, or Java project)

Installation at a Glance

  1. No installation needed - uvx will handle it automatically!
  2. Edit claude_desktop_config.json: 
    {
  "mcpServers": {
    "code-scalpel": {
      "command": "uvx",
      "args": ["codescalpel", "mcp"]
    }
  }
}
  3. Restart Claude Desktop

Alternative: pip install 

pip install codescalpel
Then use "command": "codescalpel" in config instead of uvx

  1. Install Claude Code if you haven't already
  2. Register Code Scalpel — one command, no config file editing: 
    claude mcp add codescalpel uvx codescalpel mcp
  3. Verify: claude mcp list — you should see codescalpel: uvx codescalpel mcp
  4. Launch in your project: cd /your/project && claude
  1. No installation needed - uvx will handle it automatically!
  2. Add to VS Code settings.json: 
    {
  "servers": {
    "code-scalpel": {
      "command": "uvx",
      "args": ["codescalpel", "mcp"],
      "env": {
        "CODE_SCALPEL_LICENSE_PATH": "${workspaceFolder}/.code-scalpel/license/"
      }
    }
  }
}
  3. Reload VS Code window

Alternative: pip install 

pip install codescalpel
Then use "command": "codescalpel" in config instead of uvx

  1. No installation needed - uvx will handle it automatically!
  2. Configure in Cursor settings under MCP Servers
  3. Add server with command uvx codescalpel mcp

Alternative: pip install 

pip install codescalpel
Then use command codescalpel mcp in Cursor settings

pip install codescalpel
codescalpel --help

Then configure your AI assistant with: 

{
  "command": "codescalpel",
  "args": ["mcp"]
}


docker run -v $(pwd):/workspace ghcr.io/your-org/code-scalpel:latest

Detailed installation guides

Verify Installation

After installation, verify Code Scalpel is working by asking your AI assistant:

"Use Code Scalpel to analyze the structure of this Python file."

You should see the AI invoke the analyze_code tool and return structured information about functions, classes, and imports.

What's Next?

  1. Quick Start - Get running in 5 minutes
  2. First Analysis - Understand tool responses
  3. Tutorials - Learn common workflows
  4. Tool Reference - Explore all 23 tools