FoundryVTT MCP Server

A Model Context Protocol (MCP) server that integrates with FoundryVTT, allowing AI assistants to interact with your tabletop gaming sessions. Query actors, roll dice, generate content, and manage your game world through natural language.

Features

Core Functionality

🎲 Dice Rolling - Roll dice with standard RPG notation
🔍 Data Querying - Search actors, items, scenes, and journal entries
📊 Game State - Access current scene, combat status, and world information
🎭 Content Generation - Generate NPCs, loot, and random encounters
📝 Rule Lookup - Query game rules and mechanical information

Real-time Integration

🔄 Live Updates - WebSocket connection for real-time game state
⚔️ Combat Management - Track initiative and combat state
👥 User Awareness - See who's online and their status

AI-Powered Features

🧠 Tactical Suggestions - Get combat advice and strategy tips
🎪 Story Assistance - Generate plot hooks and narrative elements
🎨 World Building - Create locations, NPCs, and quests on demand

Installation

Prerequisites

Node.js 18+
FoundryVTT server running and accessible
MCP-compatible AI client (Claude Desktop, etc.)

Quick Setup (Recommended)

🧙‍♂️ Interactive Setup Wizard:

git clone <repository-url>
cd foundry-mcp-server
npm install
npm run setup-wizard

The setup wizard will:

Automatically detect your FoundryVTT server
Test connectivity and authentication
Generate your .env configuration file
Validate the complete setup

Manual Setup

Clone and install:

git clone <repository-url>
cd foundry-mcp-server
npm install

Configure environment:

cp .env.example .env
# Edit .env with your FoundryVTT details

Required environment variables:

FOUNDRY_URL=http://localhost:30000
FOUNDRY_API_KEY=your_api_key_here
# OR use username/password:
FOUNDRY_USERNAME=your_username
FOUNDRY_PASSWORD=your_password

Test and start:

npm run test-connection  # Verify setup
npm run build
npm start

Development Mode

npm run dev

FoundryVTT Configuration

The MCP server supports two secure, local-only authentication methods:

Option 1: Local REST API Module (🔒 Recommended)

Benefits:

✅ 100% Local - No external dependencies or third-party services
✅ Maximum Privacy - Your game data never leaves your network
✅ Full Control - You own and manage all authentication
✅ Better Performance - Direct local API access
✅ Complete API Access - Full access to all FoundryVTT features

Setup:

Install the Foundry Local REST API module:
- In FoundryVTT: Setup → Add-on Modules → Install Module
- Paste: https://github.com/laurigates/foundryvtt-mcp/releases/latest/download/module.json
Enable the module in your world
Go to Settings → Configure Settings → Module Settings
Find "Foundry Local REST API" and check "Enable REST API"
Copy the generated API Key
Add to your .env file:
FOUNDRY_URL=http://localhost:30000 FOUNDRY_API_KEY=your_local_api_key_here

Option 2: Username/Password (Fallback)

Use when: Local REST API module is not available or for simple setups.

Limitations: Some advanced features may not work properly.

Ensure your FoundryVTT user has appropriate permissions
Add credentials to .env file:
FOUNDRY_URL=http://localhost:30000 FOUNDRY_USERNAME=your_username FOUNDRY_PASSWORD=your_password

Comparison Table

Feature	Local REST API Module	Username/Password
Privacy	✅ 100% Local	✅ 100% Local
Security	✅ API Key auth	⚠️ Password auth
Performance	✅ Direct API access	⚠️ WebSocket only
Features	✅ Complete API access	❌ Limited functionality
Setup	⚠️ Module install required	✅ Simple credentials
Reliability	✅ Stable API	⚠️ Connection dependent

Required Permissions (All Methods)

Your FoundryVTT user needs these permissions:

View actors, items, scenes, and journals
Create and modify journal entries (for content generation)
Access compendium data
Use dice rolling API

Usage

Basic Queries

Ask your AI assistant things like:

Dice Rolling:

"Roll 1d20+5 for an attack roll"
"Roll 4d6 drop lowest for ability scores"
"Roll 2d10+3 for damage"

Game Data:

"Show me all the NPCs in this scene"
"Find magic weapons in the party's inventory"
"What's the current combat initiative order?"
"Search for healing potions"

Content Generation:

"Generate a random NPC merchant"
"Create loot for a CR 5 encounter"
"Generate a tavern with NPCs and plot hooks"

Advanced Features

Rule Lookups:

"Look up the grappling rules"
"How does the Fireball spell work?"
"What are the conditions for being frightened?"

Tactical Advice:

"Suggest tactics for fighting a dragon"
"What should our wizard do this turn?"
"Analyze this combat encounter"

World Building:

"Create a mysterious forest location"
"Generate a side quest involving missing merchants"
"Design a magic item appropriate for level 8 characters"

Available Tools

Data Access

search_actors - Find characters, NPCs, monsters
search_items - Find equipment, spells, consumables
search_journals - Search notes and handouts
get_scene_info - Current scene details
get_actor_details - Detailed character information

Game Mechanics

roll_dice - Roll dice with any formula
update_actor_hp - Modify character health
get_combat_status - Combat state and initiative
lookup_rule - Game rules and spell descriptions

Content Generation

generate_npc - Create random NPCs
generate_loot - Create treasure appropriate for level
roll_table - Random encounters, events, weather
suggest_tactics - Combat advice and strategy

Diagnostics & System Health

get_system_health - Server performance and health metrics
get_recent_logs - Retrieve filtered FoundryVTT logs
search_logs - Search logs with regex patterns
diagnose_errors - Analyze errors with troubleshooting suggestions

Available Resources

The server exposes these FoundryVTT resources:

foundry://world/info - World and campaign information
foundry://world/actors - All actors in the world
foundry://scene/current - Current active scene
foundry://combat/current - Active combat state
foundry://compendium/spells - Spell database
foundry://compendium/monsters - Monster database

Configuration

Server Settings

Edit .env to customize:

# Logging
LOG_LEVEL=info  # debug, info, warn, error

# Performance
FOUNDRY_TIMEOUT=10000      # Request timeout (ms)
FOUNDRY_RETRY_ATTEMPTS=3   # Retry failed requests
CACHE_TTL_SECONDS=300      # Cache data for 5 minutes

Security

Use API keys instead of passwords when possible
Limit FoundryVTT user permissions to minimum required
Run server on internal network only
Monitor logs for suspicious activity

Diagnostics & Troubleshooting

Built-in Diagnostics

The server includes comprehensive diagnostic tools to help troubleshoot connection and performance issues:

Connection Testing:

# Test complete MCP connection and functionality
npm run test-connection

# Clean build and test setup
npm run setup

Diagnostic Tools (via AI assistant):

System Health: "Get the FoundryVTT system health status"
Error Analysis: "Diagnose recent errors and provide recommendations"
Log Search: "Search logs for 'connection' patterns in the last hour"
Recent Issues: "Show me recent error logs"

Advanced Diagnostics

When using the Local REST API module, you get access to advanced diagnostic features:

🔍 Real-time Log Analysis - Monitor FoundryVTT console output and notifications
📊 System Health Metrics - Server performance, memory usage, and client connections
🎯 Error Pattern Recognition - Automatic detection of common issues
💡 Smart Suggestions - Context-aware troubleshooting recommendations
📈 Performance Monitoring - Track server uptime and response times

Connection Issues

# Test FoundryVTT connection
curl http://localhost:30000/api/status

# Check server logs
npm run dev  # Shows detailed logging

Common Problems

"Failed to connect to FoundryVTT"

Verify FOUNDRY_URL is correct
Check if FoundryVTT is running
Ensure API access is enabled

"Authentication failed"

Verify API key or username/password
Check user permissions in FoundryVTT
Ensure user is not banned/restricted

"Tool not found" errors

Update to latest server version
Check tool name spelling
Review available tools in logs

Development

Project Structure

src/
├── config/           # Configuration management
├── foundry/          # FoundryVTT client and types
├── tools/            # MCP tool definitions
├── resources/        # MCP resource definitions
├── utils/            # Utilities and logging
└── index.ts          # Main server entry point

Adding New Tools

Define tool schema in src/tools/index.ts
Add handler method in src/index.ts
Implement FoundryVTT API calls in src/foundry/client.ts
Add TypeScript types in src/foundry/types.ts
Test with your AI assistant

Testing

# Run tests
npm test

# Run with coverage
npm run test:coverage

# Lint code
npm run lint

Building

# Development build
npm run build

# Clean build
npm run clean && npm run build

API Reference

Environment Variables

Variable	Required	Description	Default
`FOUNDRY_URL`	✅	FoundryVTT server URL	-
`FOUNDRY_API_KEY`	⭐	API key for authentication	-
`FOUNDRY_USERNAME`	⭐	Username (if no API key)	-
`FOUNDRY_PASSWORD`	⭐	Password (if no API key)	-
`LOG_LEVEL`	❌	Logging verbosity	`info`
`NODE_ENV`	❌	Environment mode	`development`
`FOUNDRY_TIMEOUT`	❌	Request timeout (ms)	`10000`
`FOUNDRY_RETRY_ATTEMPTS`	❌	Retry failed requests	`3`
`CACHE_TTL_SECONDS`	❌	Cache duration	`300`

⭐ Either API key OR username/password required

Tool Schemas

roll_dice

{
  "formula": "1d20+5",
  "reason": "Attack roll against goblin"
}

search_actors

{
  "query": "goblin",
  "type": "npc",
  "limit": 10
}

generate_npc

{
  "race": "human",
  "level": 5,
  "role": "merchant",
  "alignment": "neutral good"
}

Integration Examples

Claude Desktop Configuration

Add to your Claude Desktop MCP settings:

{
  "mcpServers": {
    "foundry": {
      "command": "node",
      "args": ["/path/to/foundry-mcp-server/dist/index.js"],
      "env": {
        "FOUNDRY_URL": "http://localhost:30000",
        "FOUNDRY_API_KEY": "your_api_key_here"
      }
    }
  }
}

Custom MCP Client

import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { StdioClientTransport } from "@modelcontextprotocol/sdk/client/stdio.js";

const transport = new StdioClientTransport({
  command: "node",
  args: ["./dist/index.js"],
});

const client = new Client(
  {
    name: "foundry-client",
    version: "1.0.0",
  },
  {
    capabilities: {},
  },
);

await client.connect(transport);

// Roll dice
const result = await client.request({
  method: "tools/call",
  params: {
    name: "roll_dice",
    arguments: {
      formula: "1d20+5",
      reason: "Initiative roll",
    },
  },
});

Roadmap

Version 0.2.0

Combat management tools (start/end combat, advance initiative)
Token manipulation (move, update status effects)
Scene navigation and switching
Playlist controls and ambient audio

Version 0.3.0

Character sheet editing (level up, add equipment)
Journal entry creation and editing
Macro execution and management
Advanced content generation (dungeons, NPCs with full stats)

Version 1.0.0

Multi-world support
User permission management
Webhook support for external triggers
Performance optimization and caching
Full test coverage
Docker deployment

Documentation

Complete API documentation is available in the docs/ directory, auto-generated from TypeScript source code and JSDoc comments.

📖 Viewing Documentation

Local development:

npm run docs        # Generate documentation
npm run docs:serve  # Generate and serve locally

Online: Browse the docs/ folder in this repository or visit the GitHub Pages site (if enabled).

📚 What's Documented

FoundryClient API - Complete client documentation with examples
TypeScript Interfaces - All data structures and type definitions
Configuration - Environment variables and setup options
Utilities - Helper functions and logging
Usage Examples - Code samples for common operations

The documentation is automatically updated via GitHub Actions when source code changes.

Contributing

Fork the repository
Create a feature branch: git checkout -b feature/amazing-feature
Make your changes and add tests
Commit: git commit -m 'Add amazing feature'
Push: git push origin feature/amazing-feature
Open a Pull Request

Code Style

Use TypeScript strict mode
Follow existing naming conventions
Add JSDoc comments for public APIs
Write tests for new functionality
Use meaningful commit messages

License

MIT License - see LICENSE file for details.

Troubleshooting

🔍 Quick Diagnostics

npm run test-connection      # Test FoundryVTT connectivity
npm run setup-wizard        # Re-run interactive setup

🏥 Health Check

Use the get_health_status MCP tool for comprehensive diagnostics, or check server logs during startup for detailed status information.

📚 Common Issues

Connection refused: Ensure FoundryVTT is running on the configured port
Authentication failed: Verify API key or username/password in .env
Empty search results: Install and enable the "Foundry Local REST API" module
Limited functionality: REST API module required for full features

📖 Detailed troubleshooting guide: TROUBLESHOOTING.md

Support

Issues: GitHub Issues for bugs and feature requests
Discord: FoundryVTT Discord #api-development
Documentation: FoundryVTT API Docs
Troubleshooting: TROUBLESHOOTING.md

Acknowledgments

FoundryVTT team for the excellent VTT platform
Anthropic for the Model Context Protocol
The tabletop gaming community for inspiration and feedback

Happy Gaming! 🎲

This server cannot be installed

security - not tested

license - permissive license

quality - not tested

How are these scores calculated?

hybrid server

The server is able to function both locally and remotely, depending on the configuration or use case.

A Model Context Protocol server that integrates with FoundryVTT, allowing AI assistants to interact with tabletop gaming sessions through natural language to query actors, roll dice, generate content, and manage game worlds.

Related MCP Servers

VRChat MCP Server
sawa-zen
A
security
A
license
A
quality
A Model Context Protocol server that allows AI assistants to interact with the VRChat API, enabling retrieval of user information, friends lists, and other VRChat data through a standardized interface.
Last updated -
7
22
39
TypeScript
MIT License
Godot MCP
Coding-Solo
A
security
A
license
A
quality
A Model Context Protocol server that enables AI assistants to interact with the Godot game engine, allowing them to launch the editor, run projects, capture debug output, and control project execution.
Last updated -
14
651
JavaScript
MIT License
SupaUI MCP Server
buoooou
A
security
F
license
A
quality
A Model Context Protocol server that enables AI agents to generate, fetch, and manage UI components through natural language interactions.
Last updated -
3
255
4
TypeScript
Foundry MCP Project
lhemerly
-
security
A
license
-
quality
A Model Context Protocol server that enables interaction with Foundry tools through natural language, allowing users to create projects, build contracts, run tests, and manage Ethereum development environments.
Last updated -
1
Python
MIT License

View all related MCP servers