mcpdirectory.dev
9,129 servers · updated 23 Sept

Codescene MCP server

community Active last commit 20 d
{
  "mcpServers": {
    "codescene": {
      "command": "npx",
      "args": [
        "-y",
        "@codescene/codehealth-mcp@latest"
      ],
      "env": {
        "CS_ACCESS_TOKEN": "<cs_access_token>"
      }
    }
  }
}
file .mcp.json (in your project root)install npx -y @codescene/codehealth-mcpversion 0.3.1

What it does

An MCP server that exposes CodeScene's Code Health analysis to AI assistants, enabling them to access code quality insights, maintainability issues, and technical debt information from your codebase. It runs locally and integrates with GitHub Copilot, Cursor, Claude, and other AI tools.

  • Get AI assistants to flag code quality issues like complexity and low cohesion before they generate refactoring suggestions.
  • Retrieve Code Health metrics for specific files or functions to provide AI with context about technical debt areas.
  • Integrate code quality insights into AI-driven development workflows to improve the quality of generated or refactored code.

Tools this server exposes

ToolDoes
get_code_healthRetrieve Code Health analysis and metrics for a file or function
analyze_maintainabilityGet detailed maintainability issues including complexity, nesting depth, and cohesion problems
check_technical_debtIdentify technical debt areas in the codebase

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.