This project enables Large Language Models (LLMs) like Claude to interactively debug code within VS Code. It targets developers who want AI-assisted debugging, offering a language-agnostic approach by leveraging VS Code's debugging capabilities and the Model Context Protocol (MCP). The primary benefit is a more efficient and guided debugging experience.

How It Works

The extension acts as an MCP server, bridging VS Code's debugger with an LLM client. It exposes debugger functionality (like setting breakpoints, evaluating expressions, and stepping through code) via MCP. This allows LLMs to receive real-time execution context and suggest debugging actions, facilitating interactive problem-solving. The language-agnostic nature stems from relying on VS Code's existing debugger adapters and launch.json configurations.

Quick Start & Requirements

• Install: Download from VS Code Marketplace or install from VSIX.
• Prerequisites: VS Code, Node.js, and an LLM client supporting MCP (e.g., Claude Desktop, Continue, Cursor). A valid .vscode/launch.json configuration is required for the target project.
• Setup: Requires copying a server path and configuring the LLM client's claude_desktop_config.json or equivalent.
• Docs: https://docs.continue.dev/customize/tools

Highlighted Details

• Enables LLMs to interactively debug and evaluate expressions.
• Language-agnostic, dependent on VS Code debugger support.
• Demonstrates integration with Claude Desktop and Continue.
• Supports both stdio and SSE transport for MCP.

Maintenance & Community

• Open to contributions via pull requests and issues.
• Ideas for future development include ripgrep integration and a "fix" tool.

Licensing & Compatibility

• License not explicitly stated in the README.
• Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The project is in active development, with features like conditional breakpoints and an automated "fix" tool still pending. The README does not specify the license, which could impact commercial adoption.