Plex Assistant MCP

🎬 MCP (Model Context Protocol) server for managing your Plex library through AI assistants like Claude.

Plex Assistant MCP enables you to control Plex directly from Claude Desktop or any MCP-compatible client. Search for content, manage playlists, mark items as watched, and get library statistics using natural language!

Features

• 🔍 Search - Find movies, shows, music, and photos across all libraries
• 📊 Statistics - View library overview and content breakdown
• 👥 Live Sessions - See who's watching what right now
• 📋 Playlists - Create, view, and manage playlists
• ✅ Watchlist - Mark content as watched/unwatched
• 🏷️ Collections - Organize content into custom collections
• 📡 Server Info - Get Plex server details and status
• 🎯 Advanced Search - Filter by library, media type, and more

Prerequisites

• Python ≥ 3.12
• uv - Fast Python package manager (Install here)
• Plex Server - Running instance with admin access
• Claude Desktop - Latest version (or any MCP-compatible client)

Quick Start

1. Clone or Create Project

# Create project directory
mkdir plex-assistant-mcp
cd plex-assistant-mcp

# Or clone if you have a repo
git clone https://github.com/yourusername/plex-assistant-mcp.git
cd plex-assistant-mcp

2. Initialize with uv

# Initialize Python project
uv init

# Install dependencies
uv sync

3. Configure Plex Connection

Create a .env file at the project root:

PLEX_URL=http://your-ip:32400
PLEX_TOKEN=your-plex-auth-token

How to Find Your Plex Token

Method 1: Using PlexAPI (Automatic)

python3 -c "from plexapi.myplex import MyPlexAccount; account = MyPlexAccount(username='your-email', password='your-password'); print(account.authenticationToken)"

Method 2: Browser DevTools

1. Open https://app.plex.tv/desktop/
2. Open DevTools (F12) → Network tab
3. Look for any request to plex.tv
4. Find header X-Plex-Token - copy that value

Method 3: Plex Settings

1. Open Plex Web App
2. Settings → Your Account → Scroll to "Authorized Applications"
3. Generate a token in your Plex account settings

4. Configure Claude Desktop

Edit your Claude Desktop configuration file:

Windows:

%APPDATA%\Claude\claude_desktop_config.json

macOS:

~/Library/Application Support/Claude/claude_desktop_config.json

Linux:

~/.config/Claude/claude_desktop_config.json

Add the following MCP server configuration:

{
  "mcpServers": {
    "plex-assistant": {
      "command": "uv",
      "args": [
        "--directory",
        "C:/Users/YourUsername/Documents/plex-assistant-mcp",
        "run",
        "src/plex_assistant_mcp/main.py"
      ],
      "env": {
        "PLEX_URL": "http://localhost:32400",
        "PLEX_TOKEN": "your-actual-token-here"
      }
    }
  }
}

Important:

• Update the --directory path to match your installation
• Replace PLEX_URL with your actual Plex server address
• Replace PLEX_TOKEN with your actual authentication token

5. Restart Claude Desktop

Close and reopen Claude Desktop. The Plex tools will now be available!

Usage Examples

Search for Content

"Search for The Matrix in my library"
"Find all movies from 2024"
"Show me TV shows with drama in the title"

Library Management

"How many items do I have in my Plex library?"
"Show me my movie library statistics"
"What libraries do I have?"

Active Sessions

"Who's watching something right now?"
"Show me current Plex sessions"

Watchlist & Marking

"Mark The Dark Knight as watched"
"Add Oppenheimer to my watchlist"
"Mark this show as unwatched"

Playlists

"Create a playlist called Favorites"
"Show me all my playlists"
"Get my available playlists"

Collections

"Add this to my Sci-Fi collection"
"Mark Inception as part of my Best Movies collection"

Available Tools

Local Development & Testing

Run the MCP Server Directly

export PLEX_URL="http://your-ip:32400"
export PLEX_TOKEN="your-token"
uv run src/plex_assistant_mcp/main.py

The server will start and wait for MCP commands via stdio.

Test Connection

python3 << 'EOF'
from src.plex_assistant_mcp.plex_client import PlexClient
import os

client = PlexClient(
    os.getenv("PLEX_URL", "http://localhost:32400"),
    os.getenv("PLEX_TOKEN", "")
)
print("Connection test:", client.test_connection())
print("Server info:", client.get_server_info())
EOF

Project Structure

plex-assistant-mcp/
├── src/
│   └── plex_assistant_mcp/
│       ├── __init__.py           # Package initialization
│       ├── main.py               # MCP server entry point
│       ├── plex_client.py         # Plex API wrapper
│       └── tools.py              # MCP tool definitions & handlers
├── pyproject.toml                # Project configuration
├── requirements.txt              # Python dependencies
├── .env.example                  # Example environment variables
└── README.md                     # This file

Troubleshooting

Error: PLEX_TOKEN environment variable not set

Solution: Make sure your .env file exists and contains:

PLEX_URL=http://your-ip:32400
PLEX_TOKEN=your-token

Error: Connection refused or Unable to connect

Solutions:

• Verify Plex is running on your server
• Check that the PLEX_URL is correct (including IP and port)
• Test the URL in your browser: http://your-ip:32400
• Check firewall rules on your server
• Ensure Claude Desktop can reach your Plex server

Error: HTTP 401: Unauthorized

Solutions:

• Verify your PLEX_TOKEN is correct
• Generate a new token if the old one expired
• Ensure you're using the correct authentication token format

Error: No results found for search

Solutions:

• Try different spelling or keywords
• Use exact titles if possible
• Check that content exists in your Plex library
• Verify library is properly indexed

Claude Desktop Shows No Plex Tools

Solutions:

• Verify the configuration file is properly formatted JSON
• Restart Claude Desktop completely (not just refresh)
• Check the MCP server logs in Claude Desktop settings
• Verify the --directory path is correct
• Test connection manually to ensure Plex is reachable

API Reference

PlexClient Methods

# Connection
test_connection() -> bool

# Server & Libraries
get_server_info() -> Dict[str, Any]
get_libraries() -> List[Dict[str, Any]]
get_library_statistics() -> Dict[str, Any]

# Search & Discovery
search(query: str, limit: int = 20) -> List[Dict[str, Any]]
search_in_library(query: str, library_key: str, media_type: str = "") -> List[Dict[str, Any]]
find_item(title: str) -> Optional[Dict[str, Any]]

# Sessions
get_currently_playing() -> List[Dict[str, Any]]

# Playlists
get_playlists() -> List[Dict[str, Any]]
create_playlist(title: str, items: Optional[List] = None, description: str = "") -> Dict[str, Any]

# Content Management
set_watched(item_key: str) -> bool
set_unwatched(item_key: str) -> bool
add_to_collection(item_key: str, collection_name: str) -> bool
remove_from_collection(item_key: str, collection_name: str) -> bool

Resources

• Plex Official Site
• Plex API Documentation
• PlexAPI Python Library
• MCP Protocol
• MCP Python SDK
• Claude Documentation

Contributing

Contributions are welcome! Feel free to:

• Report bugs and issues
• Suggest new features
• Submit pull requests
• Improve documentation

License

This project is open source and available under the MIT License.

Disclaimer

This project is not officially affiliated with Plex or Anthropic. Use at your own risk and ensure you have proper authorization to access your Plex server.

Made with ❤️ for Plex enthusiasts who want AI-powered library management