MCP Server

What is MCP?

The Model Context Protocol (MCP) is an open standard that enables AI assistants to connect with external data sources and tools. Thinky Face provides an MCP server that allows AI assistants like Claude, ChatGPT, and others to discover and retrieve prompts from our prompt library.

Authentication Required

To use the Thinky Face MCP server, you'll need an API token for authentication. API tokens are completely free and only require a Thinky Face account.

Get Your Free API Token

1. Create a free account (if you don't have one already)
2. Visit your API Tokens page in settings
3. Generate a new API token
4. Copy the token and add it to your MCP client configuration

Server URL

Connect your AI assistant to our MCP server at:

https://mcp.thinkyface.com

Available Tools

Our MCP server provides two tools for working with prompts:

find_prompt

Search for prompts by keywords. Returns a list of matching public prompts with their names, descriptions, and IDs.

Parameters:

• query (required): Search query to find prompts. Searches across prompt names, descriptions, and file contents.
• limit (optional): Maximum number of results to return (default: 20, max: 100)
• offset (optional): Number of results to skip for pagination (default: 0)

get_prompt

Get full details of a prompt by ID. Returns the prompt name, description, and all associated prompt files with their contents.

Parameters:

• id (required): The prompt ID (prefix_id format, e.g., 'prompt_abc123')

How to Install

Cursor

Add the following to ~/.cursor/mcp.json, replacing YOUR_API_TOKEN with your actual API token:

{
  "mcpServers": {
    "thinkyface": {
      "url": "https://mcp.thinkyface.com",
      "headers": {
        "THINKYFACE_API_KEY": "YOUR_API_TOKEN"
      }
    }
  }
}

After updating the configuration, restart Cursor for the changes to take effect.

Claude Code

Run this command in your terminal, replacing YOUR_API_TOKEN with your actual API token:

claude mcp add --transport http thinkyface https://mcp.thinkyface.com \
  --header "THINKYFACE_API_KEY: YOUR_API_TOKEN"

Or manually add to your Claude Code configuration:

{
  "mcpServers": {
    "thinkyface": {
      "type": "url",
      "url": "https://mcp.thinkyface.com",
      "headers": {
        "THINKYFACE_API_KEY": "YOUR_API_TOKEN"
      }
    }
  }
}

You can add this configuration to:

• Project-level: Create or edit .mcp.json in your project root
• User-level: Edit ~/.claude/mcp.json for global access

Claude Desktop

Method 1: Using the UI (Recommended)

1. Open Claude Desktop
2. Navigate to Settings > Connectors > Add Custom Connector
3. Enter name as Thinky Face
4. Enter URL as https://mcp.thinkyface.com
5. Add a header with key THINKYFACE_API_KEY and value YOUR_API_TOKEN (replace with your API token)
6. Click "Add"

Method 2: Manual Configuration

Add the following to your Claude Desktop configuration file, replacing YOUR_API_TOKEN with your actual token:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "thinkyface": {
      "type": "url",
      "url": "https://mcp.thinkyface.com",
      "headers": {
        "THINKYFACE_API_KEY": "YOUR_API_TOKEN"
      }
    }
  }
}

After updating the configuration, restart Claude Desktop for the changes to take effect.

Gemini CLI

Edit ~/.gemini/settings.json, replacing YOUR_API_TOKEN with your actual API token:

{
  "mcpServers": {
    "thinkyface": {
      "httpUrl": "https://mcp.thinkyface.com",
      "headers": {
        "Accept": "application/json, text/event-stream",
        "THINKYFACE_API_KEY": "YOUR_API_TOKEN"
      }
    }
  }
}

After updating the configuration, restart the Gemini CLI for the changes to take effect.

VS Code

Add the following to your VS Code settings (File > Preferences > Settings, then search for "MCP"), replacing YOUR_API_TOKEN with your actual API token:

"mcp": {
  "servers": {
    "thinkyface": {
      "type": "http",
      "url": "https://mcp.thinkyface.com",
      "headers": {
        "THINKYFACE_API_KEY": "YOUR_API_TOKEN"
      }
    }
  }
}

Or edit your settings.json file directly and add the configuration above.

After updating the configuration, reload VS Code for the changes to take effect.

Firebender

Add the following to your project's firebender.json file, replacing YOUR_API_TOKEN with your actual API token:

{
  "mcpServers": {
    "thinkyface": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/mcp-remote", "https://mcp.thinkyface.com"],
      "env": {
        "HTTP_THINKYFACE_API_KEY": "YOUR_API_TOKEN"
      }
    }
  }
}

After updating the configuration, access your MCP settings through: Plugin Settings → Firebender Settings → MCP to enable the server.

Example Workflows

Once connected, you can ask your AI assistant to:

• Search for prompts: "Find prompts about code review"
• Get prompt details: "Get the full content of prompt_abc123"
• Use prompts in your work: "Find a prompt for writing API documentation and use it to document my code"

Access and Limitations

• Authentication required: All requests must include a valid API token
• Only public prompts and prompts you have accesss to via your accounts are accessible through the MCP server
• Search is case-insensitive and matches partial text
• Rate limiting may apply to prevent abuse
• API tokens are free and can be managed from your API Tokens page

Need Help?

If you have questions or run into issues connecting to our MCP server:

• Check that your API token is correctly configured
• Ensure the THINKYFACE_API_KEY header is properly set to your API token
• Contact us at support@thinkyface.com