AI Workspace MCP Server

AI Workspace MCP Server

A Model Context Protocol (MCP) server that provides AI with a secure workspace for file management and Python script execution. Designed to run on Vercel as a serverless function.

Features

File Management Tools

• create_file - Create new files with content
• read_file - Read file contents
• update_file - Update existing files
• delete_file - Delete files
• list_files - List files and directories
• create_directory - Create new directories

Code Execution

• execute_python - Execute Python scripts with arguments (30-second timeout)

Setup on Vercel

1. Install Vercel CLI (Optional)

npm install -g vercel

2. Project Structure

Your project should look like this:

ai-workspace-mcp/
├── api/
│   └── mcp.py          # Serverless function
├── vercel.json         # Vercel configuration
├── requirements.txt    # Python dependencies
└── README.md          # This file

3. Deploy to Vercel

Option A: Deploy via Vercel Dashboard

1. Go to vercel.com
2. Click "Add New" → "Project"
3. Import your Git repository (or upload files)
4. Vercel will auto-detect Python and deploy

Option B: Deploy via CLI

# Login to Vercel
vercel login

# Deploy
vercel

# Deploy to production
vercel --prod

4. Get Your Deployment URL

After deployment, Vercel will give you a URL like: https://your-project-name.vercel.app

API Endpoints

Once deployed, your server will have these endpoints:

GET /

Returns server information and status

curl https://your-project.vercel.app/

GET /health

Health check endpoint

curl https://your-project.vercel.app/health

GET /tools

List all available tools

curl https://your-project.vercel.app/tools

POST /execute

Execute a tool

curl -X POST https://your-project.vercel.app/execute \
  -H "Content-Type: application/json" \
  -d '{
    "tool": "create_file",
    "arguments": {
      "filepath": "hello.py",
      "content": "print(\"Hello World!\")"
    }
  }'

Using with AI Clients

Claude Desktop Configuration

Add this to your Claude Desktop config:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "ai-workspace": {
      "command": "curl",
      "args": [
        "-X", "POST",
        "https://your-project.vercel.app/execute",
        "-H", "Content-Type: application/json",
        "-d", "@-"
      ]
    }
  }
}

Using the API Directly

You can integrate this with any AI that supports HTTP tool calling:

import requests

# Create a file
response = requests.post(
    "https://your-project.vercel.app/execute",
    json={
        "tool": "create_file",
        "arguments": {
            "filepath": "script.py",
            "content": "print('Hello from AI!')"
        }
    }
)
print(response.json())

# Execute the file
response = requests.post(
    "https://your-project.vercel.app/execute",
    json={
        "tool": "execute_python",
        "arguments": {
            "filepath": "script.py"
        }
    }
)
print(response.json())

Security Features

• Sandboxed Workspace: All file operations are restricted to /tmp/workspace
• Path Validation: Prevents directory traversal attacks
• Execution Timeout: Python scripts are limited to 30 seconds
• CORS Enabled: Allows cross-origin requests
• Serverless Isolation: Each request runs in an isolated environment

Tool Examples

Create and Execute a Python Script

# Create a file
curl -X POST https://your-project.vercel.app/execute \
  -H "Content-Type: application/json" \
  -d '{
    "tool": "create_file",
    "arguments": {
      "filepath": "hello.py",
      "content": "print(\"Hello from Vercel!\")"
    }
  }'

# Execute it
curl -X POST https://your-project.vercel.app/execute \
  -H "Content-Type: application/json" \
  -d '{
    "tool": "execute_python",
    "arguments": {
      "filepath": "hello.py"
    }
  }'

List Files

curl -X POST https://your-project.vercel.app/execute \
  -H "Content-Type: application/json" \
  -d '{
    "tool": "list_files",
    "arguments": {}
  }'

Create Directory Structure

curl -X POST https://your-project.vercel.app/execute \
  -H "Content-Type: application/json" \
  -d '{
    "tool": "create_directory",
    "arguments": {
      "dirpath": "scripts"
    }
  }'

Response Format

All tool executions return JSON:

Success Response:

{
  "success": true,
  "message": "Successfully created file: hello.py\nSize: 26 bytes"
}

Error Response:

{
  "success": false,
  "error": "File not found: nonexistent.py"
}

Execute Python Response:

{
  "success": true,
  "exit_code": 0,
  "stdout": "Hello from Vercel!\n",
  "stderr": ""
}

Important Notes

Vercel Limitations