Debugmcp
Gift your VS Code agent a real debugger: breakpoints, stepping, inspection.
What is Debugmcp?
Debugmcp is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to gift your vs code agent a real debugger: breakpoints, stepping, inspection.
Gift your VS Code agent a real debugger: breakpoints, stepping, inspection.
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Gift your VS Code agent a real debugger: breakpoints, steppi
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx debugmcpConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Debugmcp
DebugMCP is a VS Code extension that exposes a full interactive debugger to AI coding agents through the Model Context Protocol. It provides 14 tools covering breakpoint management, step execution (over, into, out), variable inspection, expression evaluation, and session control — giving agents the same runtime introspection capabilities a human developer has in the VS Code debugger. The extension works with every language VS Code can debug, including Python, JavaScript, TypeScript, Java, Go, Rust, C/C++, Ruby, PHP, and C#.
Prerequisites
- VS Code with the DebugMCP extension installed (marketplace ID: ozzafar.debugmcpextension)
- The appropriate VS Code debugger extension for your language (e.g., Python, Go, or Java extension packs)
- An MCP-compatible AI client such as Claude Desktop, Cline, Windsurf, or GitHub Copilot
- Node.js is not required — the extension runs a built-in HTTP server on port 3001
Install the DebugMCP VS Code extension
Open VS Code and install the extension from the marketplace. Search for 'DebugMCP' in the Extensions panel or install directly via the marketplace link.
# Install via VS Code quick open (Ctrl+P):
ext install ozzafar.debugmcpextensionVerify the server is running
Once the extension is active, it automatically starts an HTTP/MCP server on localhost:3001. No further configuration is required for basic use. Check the VS Code status bar or Output panel for confirmation.
Configure your MCP client
Add the DebugMCP server to your AI client configuration using the streamableHttp transport type. The server binds to loopback addresses only, so this configuration is safe for local development.
{
"mcpServers": {
"debugmcp": {
"type": "streamableHttp",
"url": "http://localhost:3001/mcp"
}
}
}Adjust server settings (optional)
Customize the server port, timeout, and bind address through VS Code settings if the defaults conflict with other tools.
// In VS Code settings.json:
{
"debugmcp.serverPort": 3001,
"debugmcp.timeoutInSeconds": 180,
"debugmcp.bindHost": ["127.0.0.1", "::1"]
}Start a debug session through the AI agent
Ask your AI client to start a debug session by calling the start_debugging tool with the full path to the file you want to debug. The agent can then set breakpoints, step through code, and inspect variables — all from within the chat.
Debugmcp Examples
Client configuration (Cline / Windsurf)
Configuration for Cline, Windsurf, or Roo Code using the streamableHttp transport.
{
"mcpServers": {
"debugmcp": {
"type": "streamableHttp",
"url": "http://localhost:3001/mcp"
}
}
}Prompts to try
Example prompts to use once DebugMCP is connected to your AI client.
- "Start a debug session for src/main.py and set a breakpoint at line 42"
- "Step over the current line and show me the local variable values"
- "Evaluate the expression 'user.permissions' in the current debug context"
- "List all active breakpoints in my project"
- "Step into the processRequest function and inspect what 'payload' contains"
- "Stop the debug session and clear all breakpoints"Troubleshooting Debugmcp
AI client cannot connect to http://localhost:3001/mcp
Verify the DebugMCP extension is installed and VS Code is open. The server only starts when the extension activates. If port 3001 is in use, change debugmcp.serverPort in VS Code settings and update the URL in your MCP config. GitHub Copilot users should use the 'http' transport type (not 'streamableHttp') in their config.
start_debugging tool fails with 'no debug adapter available'
DebugMCP delegates to VS Code's debug adapter infrastructure. Make sure you have installed the appropriate language extension (e.g., the Python or Go extension) and that a valid launch configuration exists in .vscode/launch.json, or provide the configurationName parameter matching an existing configuration.
Variable inspection shows empty results after hitting a breakpoint
Ensure the debug session is actually paused at a breakpoint — call get_variables_values only after start_debugging has returned and execution has halted. Pass scope: 'local' to limit results to the current stack frame, which is faster and less noisy than scope: 'all'.
Frequently Asked Questions about Debugmcp
What is Debugmcp?
Debugmcp is a Model Context Protocol (MCP) server that gift your vs code agent a real debugger: breakpoints, stepping, inspection. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Debugmcp?
Follow the installation instructions on the Debugmcp GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Debugmcp?
Debugmcp works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Debugmcp free to use?
Yes, Debugmcp is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Debugmcp Alternatives — Similar Developer Tools Servers
Looking for alternatives to Debugmcp? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Debugmcp in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Debugmcp?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.