Connect your AI agents and tools to Peerdraft Spaces using our MCP Server

MCP is a standard protocol that allows AI agents to connect to external tools and services. The Peerdraft MCP Server enables AI agents to read, write, and update documents in your Spaces, enabling automated workflows, assistants, and AI-powered tooling on top of your notes.

When to Use this

You want to connect your agents or AI tools to your Peerdraft Spaces to:

  • Enable automated workflows and assistants
  • Build AI-powered tooling on top of your notes
  • Allow agents to read, write, and update documents in your Spaces

Example use cases

  • Summarize new notes
  • Keep project docs up to date from external data
  • Build a personal knowledge assistant
  • Generate reports from your Space
  • Let agents triage and organize notes

How to connect

Endpoint

Each Space exposes an MCP endpoint:

https://www.peerdraft.app/mcp/SPACE_ID

To get your space id, open the context menu of your shared space (= folder) and select “Copy MCP URL”.

The url will look something like this:

https://www.peerdraft.app/team/SPACE_ID

You only need to replace team with mcp and you have your MCP endpoint.

Also check Spaces vs. Folders for more information on why we sometimes call them Spaces and sometimes Folders.

Configuration

Claude

Claude supports two methods for connecting to MCP servers across web, desktop, and mobile apps:

This method works for users on Free, Pro, Max, Team, and Enterprise plans. Free users are limited to one custom connector. Once added, the connector is available across Claude web, Claude Desktop, and mobile apps (iOS and Android).

  1. Navigate to Customize > Connectors in Claude
  2. Click the ”+” button next to Connectors
  3. Select “Add custom connector”
  4. Enter the connector’s name (e.g., “Peerdraft Space”)
  5. Enter your MCP endpoint URL: https://www.peerdraft.app/mcp/YOUR_SPACE_ID
  6. Click “Add”

For detailed information on how to use custom connectors with Claude, see [Get started with custom connectors using remote MCP](https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp#h_3d1a65aded

Team and Enterprise plans: Before members can use connectors, an Owner or Primary Owner must enable them for the organization at Organization settings > Connectors. Individual users then authenticate individually.

Method 2: MCP Remote Bridge (For advanced users)

For Claude Desktop only, if you are on the free plan and you already connected a custom connector, use the mcp-remote bridge. Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "my-space": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://www.peerdraft.app/mcp/YOUR_SPACE_ID"]
    }
  }
}

Using connected services: Once connected, enable the connector in your conversation by clicking the ”+” button, hovering over “Connectors,” and toggling it on. Claude will use the service when relevant to your requests.

ChatGPT

ChatGPT uses “Connectors” to connect to MCP servers. This requires a Business, Enterprise, or Education plan with Developer Mode enabled.

Prerequisites:

  1. Enable Developer Mode: Settings → Apps & Connectors → Advanced settings → Toggle Developer Mode

Create a Connector:

  1. In ChatGPT, navigate to Settings → Connectors → Create
  2. Provide the following information:
    • Connector name: A user-facing title (e.g., “Peerdraft Space”)
    • Description: Explain what the connector does and when to use it
    • Connector URL: Your MCP endpoint: https://www.peerdraft.app/mcp/YOUR_SPACE_ID
  3. Click Create

Use the Connector:

  1. Open a new chat in ChatGPT
  2. Click the + button near the message composer, then click More
  3. Select your connector from the available tools list
  4. The connector is now available in the conversation context

Connectors created in ChatGPT web are also available on ChatGPT mobile apps

Cursor

Add this to your Cursor MCP config (.cursor/mcp.json):

{
  "mcpServers": {
    "my-space": {
      "url": "https://www.peerdraft.app/mcp/YOUR_SPACE_ID"
    }
  }
}

Configuration options:

  • Project-scoped: Place .cursor/mcp.json in your project root (shared with your team)
  • Global: Place at ~/.cursor/mcp.json (applies to all projects, stays local to your machine)

For detailed information on MCP configuration in Cursor, see the official Cursor MCP documentation.

Windsurf

Add this to your Windsurf MCP config file (~/.codeium/windsurf/mcp_config.json):

{
  "mcpServers": {
    "my-space": {
      "serverUrl": "https://www.peerdraft.app/mcp/YOUR_SPACE_ID"
    }
  }
}

Important: Use serverUrl (not url).

For detailed information on MCP configuration in Windsurf, see the official Windsurf MCP documentation.

Other Tools

Many other AI clients and productivity tools support MCP. Here are some popular options:

  • LibreChat - Open source ChatGPT alternative for macOS and Windows with MCP support. Documentation | Website
  • Jan - Open source ChatGPT alternative that runs 100% offline with local AI models. Documentation | Website

What You Can Do

Once you’ve configured your AI client with the MCP endpoint, you can ask it to:

  • “Summarize my meeting notes”: The AI reads your notes and generates a concise summary
  • “Create a report from my project documents”: The AI gathers information across multiple documents and compiles a structured report
  • “Organize my research”: The AI reviews your documents and suggests categories or reorganization
  • “Update this document with new information”: The AI can modify existing documents while preserving your content
  • “Find all documents about [topic]”: The AI searches through your Space to locate relevant notes

Security

  • Each Space has a unique random ID that is unguessable and required for MCP connections
  • In the future, we will add further security measures (including authentication) to the MCP endpoint
Peerdraft Docs