Fire in da houseTop Tip:Paying $100+ per month for Perplexity, MidJourney, Runway, ChatGPT and other tools is crazy - get all your AI tools in one site starting at $15 per month with Galaxy AI Fire in da houseCheck it out free

vscode-context-mcp

MCP.Pizza Chef: vilasone455

vscode-context-mcp is an MCP server that connects AI assistants with the VSCode IDE environment. It provides real-time access to VSCode context such as active files, tabs, terminal output, and diagnostics, along with controlled filesystem and shell command operations. This enables AI models to deliver context-aware development assistance by interacting directly with the user's VSCode workspace and project files.

Use This MCP server To

Access active VSCode files and tabs for context-aware AI assistance Retrieve and analyze terminal output within VSCode for debugging Perform controlled filesystem operations like read/write and directory management Search and list project files and directories inside VSCode Execute shell commands safely within the VSCode environment Fetch diagnostic problems to help identify code issues Enable AI to edit files with line-based changes in VSCode Generate directory trees for project structure visualization

README

VSCode Context MCP

A Model Context Protocol (MCP) server that provides VSCode context and filesystem operations for AI assistants.

Overview

VSCode Context MCP creates a bridge between AI assistants and your VSCode environment. It enables AI models to:

  1. Access the current VSCode context (active files, tabs, terminal output)
  2. Perform filesystem operations with controls
  3. Execute shell commands in a controlled environment

This allows AI assistants to provide more relevant and context-aware help for software development tasks.

Features

VSCode Context Access

  • Get current project path
  • Retrieve active file contents and metadata
  • List open editor tabs
  • Access diagnostic problems (warnings/errors)
  • Retrieve terminal output
  • Access files attached to the workspace

Filesystem Operations

  • Read/write files
  • Edit files with line-based changes
  • Create directories
  • List directory contents
  • Generate directory trees
  • Search for files
  • Get file metadata
  • Move/rename files

Command Execution

  • Run shell commands with output capture
  • Controlled execution environment

⚠️ Security Warning

Be careful what you ask this server to run! In the Claude Desktop app, use Approve Once (not Allow for This Chat) so you can review each command, and use Deny if you don't trust the command. Permissions are dictated by the user running the server.

This tool does not yet implement comprehensive security measures, so treat it with caution.

Integration with AI Assistants

This MCP server is designed to work with Model Context Protocol-compatible AI assistants. It enhances their ability to:

  • Understand your code context
  • Make informed suggestions
  • Provide relevant examples
  • Help debug issues
  • Perform file operations on your behalf

1. Install the VSCode Extension

Note: Before installation, it's recommended to uninstall similar MCP servers such as filesystem MCP to avoid conflicts.

  • Download the VSIX package from here
  • Open VSCode
  • Press Ctrl+Shift+P (Windows/Linux) or Cmd+Shift+P (macOS) to open the command palette
  • Type "Extensions: Install from VSIX..." and select it
  • Navigate to the downloaded VSIX file and select it
  • Restart VSCode after installation

For more information, visit the VSCode extension repository

2. Set Up and Run the MCP Server

First, clone or download this repository. Then, install the dependencies and build the project:

cd path/to/vscode-context-mcp
npm install
npm run build

Once built, update your Claude Desktop configuration to use the MCP server:

{
  "mcpServers": {
    "vscode-context-mpc": {
      "command": "node",
      "args": [
        "C:\\path\\to\\vscode-context-mcp\\dist\\index.js"
      ]
    }
  }
}

Note: Make sure the file path in the args array matches your actual project location.

3. Test the Connection

To verify that everything is working correctly, try using this prompt in Claude:

Please check my current VSCode project and tell me what files I have open.

4. Add Files to Claude Context

You can add files to your Claude context in a few ways:

  • Press Ctrl+L (Windows/Linux) or Cmd+L (macOS) to add the current file
  • Press Ctrl+I or Cmd+I to add only the selected text
  • Open the command palette (Ctrl+Shift+P or Cmd+Shift+P) and search for:
    • "Add File to Context"
    • "Add Selection to Context"

Tool Reference

Tool Name Description
get_vscode_context Retrieve complete VSCode context information
get_attached_files Retrieve a list of files attached to the workspace
get_project_path Get current project root directory
get_current_file Get details and content of the active file
get_open_tabs List all open editor tabs
get_problems Retrieve diagnostics (errors/warnings)
get_terminal_content Get terminal output and history
run_command Execute a shell command
read_file Read contents of a file
read_multiple_files Read multiple files at once
write_file Create or overwrite a file
edit_file Make line-based edits to a file
create_directory Create directory structure
list_directory List files and directories
directory_tree Get recursive directory structure
move_file Move or rename files and directories
search_files Find files matching a pattern
get_file_info Get detailed file metadata

Roadmap

  1. Make support for multiple VSCode projects
  2. Accept/Revert functionality when writing or editing files
  3. Redesign UI

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License – see the LICENSE file for details.

vscode-context-mcp FAQ

How does vscode-context-mcp ensure safe filesystem operations?
It enforces controlled access and permissions to prevent unauthorized file changes, ensuring safe interaction with the VSCode workspace.
Can vscode-context-mcp access terminal output in real time?
Yes, it can retrieve terminal output from the VSCode integrated terminal to provide relevant context for AI models.
Does vscode-context-mcp support editing files within VSCode?
Yes, it supports line-based file edits, allowing AI assistants to modify code safely and precisely.
How does vscode-context-mcp handle multiple open editor tabs?
It can list and provide metadata for all open tabs, enabling AI to understand the current editing context.
Can vscode-context-mcp execute shell commands?
Yes, it can run shell commands in a controlled environment to assist with development workflows.
Is vscode-context-mcp limited to specific programming languages?
No, it works at the IDE and filesystem level, making it language-agnostic and useful across all VSCode projects.
How does vscode-context-mcp integrate with different LLM providers?
It is provider-agnostic and can work with models from OpenAI, Anthropic Claude, and Google Gemini by exposing VSCode context uniformly.
What kind of diagnostic information can vscode-context-mcp access?
It can retrieve warnings, errors, and other diagnostic problems reported by VSCode to help AI models identify code issues.