Codescene
@codescene-oss
The CodeScene MCP Server exposes CodeScene’s Code Health analysis as local AI-friendly tools.
Overview
What is Codescene?
The Codescene MCP Server exposes CodeScene’s Code Health analysis as local AI-friendly tools. It runs in your local environment and lets AI assistants (such as GitHub Copilot, Cursor, and Claude code) request code health insights directly from your codebase, augmenting prompts with information about code quality, maintainability, and technical debt.
How to use Codescene?
Run the server via Docker with environment variables for your CodeScene access token and the path to your code. Configure it for your AI assistant (Claude Code, Codex CLI, GitHub Copilot, Kiro, VS Code, or Amazon Q CLI) using the provided setup instructions for either CodeScene Cloud or CodeScene On-prem.
Key features of Codescene
- Exposes CodeScene’s Code Health analysis as local AI tools.
- Runs in your local environment without external data transfer.
- Supports both CodeScene Cloud and CodeScene On-prem.
- Works with multiple AI assistants (Claude Code, Copilot, etc.).
- Requires only Docker and an access token to configure.
Use cases of Codescene
- AI assistants can request code quality,
