AI Integration (MCP)

Every instance of HashCore documentation includes a built-in Model Context Protocol (MCP) server. This allows you to connect the documentation directly to your AI-powered development tools (like Claude Code, Cursor, Zed, Windsurf, or VS Code) so they can read and search our docs in real-time.

Integrating our documentation with your IDE ensures that your AI assistant has up-to-date and accurate context about HashCore APIs, configuration parameters, and product features, preventing hallucinations.

Connection URL

The MCP server is served over HTTP at the following endpoint:

How to Connect to AI Clients

1. Claude Code

You can add the documentation server to Claude Code using the following CLI command:

claude mcp add --transport http hashcore-docs https://docs.hashcore.com/mcp

2. Cursor

You can configure the server in Cursor settings under Settings > Features > MCP (add a new server with type http and the URL above), or manually create/update the .cursor/mcp.json file in your project root:

{
  "mcpServers": {
    "hashcore-docs": {
      "type": "http",
      "url": "https://docs.hashcore.com/mcp"
    }
  }
}

3. VS Code (GitHub Copilot)

Create or update the .vscode/mcp.json configuration file in your project:

{
  "servers": {
    "hashcore-docs": {
      "type": "http",
      "url": "https://docs.hashcore.com/mcp"
    }
  }
}

4. Zed

Add the context server to your global Zed configuration file (.config/zed/settings.json):

{
  "context_servers": {
    "hashcore-docs": {
      "source": "custom",
      "command": "npx",
      "args": ["mcp-remote", "https://docs.hashcore.com/mcp"],
      "env": {}
    }
  }
}

5. Windsurf

Open Windsurf, navigate to Settings > Windsurf Settings > Cascade, click Manage MCPs, select View raw config and add the following configuration to .codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "hashcore-docs": {
      "type": "http",
      "url": "https://docs.hashcore.com/mcp"
    }
  }
}

Available Tools

Once connected, your AI assistant will automatically use the following tools during your conversations:

• list-pages — Lists all available pages in the documentation with their titles, paths, and descriptions.
• get-page — Retrieves the full Markdown content of a specific documentation page.