# MCP Servers for Claude Code

A curated list of Model Context Protocol servers to enhance Claude's capabilities as a personal assistant.

## What is MCP?

Model Context Protocol (MCP) is an open standard that enables AI models to securely interact with external data sources and tools through standardized server implementations.

## Recommended MCP Servers

### 🧮 Math & Computation

#### Wolfram Alpha
Multiple implementations available for computational intelligence:

- **ricocf/mcp-wolframalpha** (Python)
  - GitHub: https://github.com/ricocf/mcp-wolframalpha
  - Features: Mathematical calculations, scientific queries, data analysis
  - Returns: Structured results in text or image formats

- **cnosuke/mcp-wolfram-alpha** (Go)
  - GitHub: https://github.com/cnosuke/mcp-wolfram-alpha
  - Features: Delegates numerical computation to Wolfram Alpha's high-precision engine

- **bhelx/wolfram-alpha**
  - MCP Hub: https://www.mcp.run/bhelx/wolfram-alpha

**Note:** Requires Wolfram Alpha API key

### 📅 Calendar Management

#### Google Calendar
- **markelaugust74/mcp-google-calendar**
  - GitHub: https://github.com/markelaugust74/mcp-google-calendar
  - Features: Create and manage calendar events through Claude
  - Integration: Google Calendar API

- **thisnick/google-calendar**
  - PulseMCP: https://www.pulsemcp.com/servers/thisnick-google-calendar
  - Features: Event creation, modification, schedule listing, free time slots

#### Apple Calendar
- Features: Natural language interaction with macOS Calendar
- Capabilities: Event creation, modification, schedule listing, finding free time slots

### ✅ Todo & Task Management

#### Todoist
- Features: Natural language task management
- Integration: Todoist API

#### Todo MCP Server (Xczer/todo-mcp-server)
- GitHub: https://github.com/Xczer/todo-mcp-server
- Technology: Node.js, TypeScript, SQLite
- Features: Tools, resources, and prompts for managing todos
- Works with: Claude and other MCP-compatible clients

#### Google Tasks
- Features: Google Tasks API integration
- Capabilities: Task creation, completion, list management

#### Things 3 (macOS)
- Integration: Things 3 to-do app on macOS
- Features: Natural language task management

### 🏠 Home Automation

#### Home Assistant
- Integration: https://www.home-assistant.io/integrations/mcp_server/
- Features: Control smart home devices, expose services as tools
- Examples: Control lights, access Google Tasks from Claude Desktop

### 🔗 Multi-App Integration

#### Rube
- Features: Connects to 500+ apps (Gmail, Slack, GitHub, Notion)
- Setup: Install in AI client, authenticate once
- Capabilities: "Send an email", "Create a task", and more

#### Routine
- Features: Interact with calendars, tasks, notes
- Integration: Routine productivity app

#### MintMCP
- Integrations: Google Calendar, Gmail, Outlook Calendar, Outlook

## Resources

### Awesome Lists
- **wong2/awesome-mcp-servers**: https://github.com/wong2/awesome-mcp-servers
- **appcypher/awesome-mcp-servers**: https://github.com/appcypher/awesome-mcp-servers
- **punkpeye/awesome-mcp-servers**: https://github.com/punkpeye/awesome-mcp-servers (with web directory)
- **TensorBlock/awesome-mcp-servers**: https://github.com/TensorBlock/awesome-mcp-servers (7260+ servers)

### Directories
- MCP.so: https://mcp.so/
- PulseMCP: https://www.pulsemcp.com/
- MCP Market: https://mcpmarket.com/
- MCP Servers Org: https://mcpservers.org/

## Priority for Installation

For personal assistant capabilities, recommended priority order:

1. **Wolfram Alpha** - Essential for math and computation
2. **Google Calendar** or **Apple Calendar** - Calendar management
3. **Todoist** or **Google Tasks** - Task management
4. **Rube** or **MintMCP** - Multi-app integration (email, etc.)

## Installed Servers

### ✅ Wolfram Alpha - INSTALLED & WORKING

**Location:** `Personal/Areas/MCP-Servers/mcp-wolframalpha`

**Installation Steps Completed:**
1. ✅ Obtained Wolfram Alpha API key from https://products.wolframalpha.com/api
2. ✅ Cloned repository: `git clone https://github.com/ricocf/mcp-wolframalpha.git`
3. ✅ Installed dependencies: `uv sync`
4. ✅ Created `.env` file with `WOLFRAM_API_KEY`
5. ✅ Configured in `.mcp.json`
6. ✅ Enabled in Claude Code settings
7. ✅ Tested and verified working

**Example Queries:**

*Mathematics:*
- "integrate x^2 from 0 to 5"
- "solve x^2 + 5x + 6 = 0"
- "derivative of sin(x) * cos(x)"
- "limit of (x^2 - 1)/(x - 1) as x approaches 1"
- "matrix determinant [[1,2],[3,4]]"

*Science & Physics:*
- "speed of light in meters per second"
- "half-life of carbon-14"
- "molecular weight of water"
- "distance from Earth to Mars"

*Unit Conversions:*
- "convert 50 miles to kilometers"
- "100 USD to EUR"
- "5 gallons to liters"
- "72 fahrenheit to celsius"

*Statistics & Data:*
- "population of Tokyo"
- "GDP of Germany"
- "average temperature in Paris in July"

*Date & Time:*
- "how many days until Christmas 2025"
- "what day of the week was January 1, 2000"
- "time difference between New York and Tokyo"

*Nutrition & Health:*
- "calories in a banana"
- "nutritional info for 100g chicken breast"

## Next Steps

- [ ] Choose preferred calendar server (Google vs Apple)
- [ ] Choose preferred todo server (Todoist vs Google Tasks vs Things 3)
- [ ] Obtain additional API keys (Google Calendar, Todoist, etc.)
- [ ] Install and configure selected servers
- [ ] Test additional integrations with Claude Code

## Installation Guide: Wolfram Alpha MCP Server

### Prerequisites
- Python 3
- `uv` package manager installed
- Wolfram Alpha API key (free tier: 2,000 calls/month)

### Step-by-Step Installation

1. **Get API Key**
   - Visit: https://products.wolframalpha.com/api
   - Sign up for free account
   - Get your App ID (API key)

2. **Clone Repository**
   ```bash
   cd Personal/Areas/MCP-Servers
   git clone https://github.com/ricocf/mcp-wolframalpha.git
   cd mcp-wolframalpha
   ```

3. **Install Dependencies**
   ```bash
   uv sync
   ```

4. **Create .env File**
   Create `.env` in the repository root:
   ```
   WOLFRAM_API_KEY=your_api_key_here
   ```

5. **Configure MCP Server**
   Create/update `.mcp.json` in your project root:
   ```json
   {
     "mcpServers": {
       "WolframAlpha": {
         "command": "uv",
         "args": [
           "--directory",
           "/absolute/path/to/mcp-wolframalpha",
           "run",
           "python",
           "src/core/server.py"
         ],
         "env": {
           "WOLFRAM_API_KEY": "your_api_key_here"
         }
       }
     }
   }
   ```

6. **Enable in Claude Code Settings**
   Add to `.claude/settings.local.json`:
   ```json
   {
     "enableAllProjectMcpServers": true
   }
   ```

7. **Restart Claude Code**
   The MCP server will be loaded on next startup.

8. **Verify Installation**
   Ask Claude to perform a calculation using Wolfram Alpha!