DollhouseMCP

A Free, Open Source MCP server for dynamic custom persona management with public a GitHub collection of personas, skills, templates, and other elements for AI models.

typescript

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "dollhousemcp-mcp-server": {
      "command": "node",
      "args": [
        "/Users/YOUR_USERNAME/mcp-servers/node_modules/@dollhousemcp/mcp-server/dist/index.js"
      ],
      "env": {
        "DOLLHOUSE_PORTFOLIO_DIR": "/Users/YOU/portfolios/personal"
      }
    }
  }
}

DollhouseMCP is an open source MCP server that lets you create, manage, and deploy customizable AI elements like personas, skills, templates, agents, and memories. It provides a local-first architecture with a comprehensive portfolio system, strong security, and a workflow for discovering, customizing, and sharing community elements across AI platforms.

How to use

You interact with DollhouseMCP through a local server that you run on your computer and a client (your MCP-enabled editor or AI workspace) that can talk to it. Start by launching the server from one of your configured methods, then use your client to browse available personas and elements, activate those you want, and manage your portfolio. You can combine elements from the community collection with your own creations, save them to your portfolio, and synchronize with GitHub for backups and sharing.

How to install

Prerequisites: you need Node.js installed on your computer. Ensure you have npm available to install packages.

# Create a directory for MCP servers
mkdir ~/mcp-servers
cd ~/mcp-servers

# Install DollhouseMCP
npm install @dollhousemcp/mcp-server

Configuration and start options

You have two common ways to run the server locally. The first uses the Node runtime with a direct path to the built index, the second uses npx to fetch the latest version on demand.

# Method 1: Local installation (STDIO)
# This runs node on the dist/index.js inside your installed package
# Replace YOUR_USERNAME with your actual home directory path

node /Users/YOUR_USERNAME/mcp-servers/node_modules/@dollhousemcp/mcp-server/dist/index.js

If you prefer to run the server via npx (always the latest), configure your client to use this command:

{
  "mcpServers": {
    "dollhousemcp_npx": {
      "command": "npx",
      "args": ["@dollhousemcp/mcp-server@latest"]
    }
  }
}

Available tools

list_portfolio_elements

View all elements across types in your local portfolio.

sync_portfolio

Synchronize your portfolio with GitHub for backup and collaboration.

upload_to_portfolio

Share elements from your local portfolio to the community collection.

download_from_portfolio

Retrieve elements from the community collection or GitHub-backed portfolio.

activate_element

Activate a specific element (persona, skill, template, agent, memory) for use.

deactivate_element

Deactivate a currently active element.

get_element_details

Show detailed information about a specific element.

browse_collection

Browse the community collection to discover available elements.

search_collection

Search the community collection for specific content or keywords.

create_element

Create a new element (persona, skill, template, agent, memory).

Built by

VeilStrat

AI signals for GTM teams