puremd MCP Server

What does “puremd” MCP Server do?

The puremd MCP (Model Context Protocol) Server is a powerful tool that connects AI assistants and agents with web content by acting as an unblocking, scraping, and search bridge. By leveraging the pure.md platform, it allows scripts, APIs, apps, and agents to access web pages in markdown format by simply prefixing any URL with pure.md/. This approach bypasses bot detection, renders JavaScript-heavy single-page apps, and converts a wide range of formats (HTML, PDF, images) into markdown. Globally cached responses ensure efficient repeated access to the same resources, reducing load on origin servers. The puremd-mcp server teaches MCP clients (like Cursor, Windsurf, and Claude Desktop) how to use these features, thus unblocking content and enabling advanced web search directly within the AI development workflow.

List of Prompts

No prompt templates are explicitly mentioned in the repository or documentation.

List of Resources

No specific resources (as defined by MCP) are described in the repository or documentation.

List of Tools

• unblock-url
  Extracts markdown from web pages without getting blocked, enabling AI clients to reliably retrieve readable content even from sites with anti-bot measures.
• search-web
  Searches the web for a given query and concatenates the results into a single markdown document, providing summarized search insights for agents and users.

Use Cases of this MCP Server

• Bypassing Bot Detection for Web Scraping
  Enables agents and LLMs to fetch content from websites that typically block automated scripts, using pure.md’s infrastructure to avoid detection.
• Web Content Conversion to Markdown
  Converts a wide variety of web resources (HTML, PDFs, images) into markdown format, standardizing input for downstream LLM processing and agent workflows.
• Web Search Integration
  Empowers AI assistants with the ability to perform web searches and access synthesized results in markdown, essential for research, summarization, and context-building.
• Global Content Caching
  Reduces server loads and ensures faster access for repeated queries by leveraging pure.md’s CDN-like caching for previously accessed resources.
• Seamless LLM Context Enrichment
  Provides a way for developers to automatically enrich LLM context windows with relevant, up-to-date web information, improving the performance of knowledge-based AI applications.

How to set it up

Windsurf

1. Ensure you have Node.js and npm installed.
2. Generate a PUREMD API key at pure.md (optional for higher rate limits).
3. Edit ./codeium/windsurf/model_config.json.
4. Add the following MCP server entry:

{
  "mcpServers": {
    "pure.md": {
      "command": "npx",
      "args": ["-y", "puremd-mcp"],
      "env": {
        "PUREMD_API_KEY": "<TOKEN>"
      }
    }
  }
}

5. Save and restart Windsurf to apply changes.

Claude

1. Ensure Node.js and npm are installed.
2. Obtain a PUREMD API key (see pure.md).
3. Edit ~/Library/Application Support/Claude/claude_desktop_config.json.
4. Insert the following configuration:

{
  "mcpServers": {
    "pure.md": {
      "command": "npx",
      "args": ["-y", "puremd-mcp"],
      "env": {
        "PUREMD_API_KEY": "<TOKEN>"
      }
    }
  }
}

5. Save and restart Claude Desktop.

Cursor

1. Install Node.js and npm.
2. Register and generate a PUREMD API key from pure.md.
3. Edit ~/.cursor/mcp.json.
4. Add the following MCP server configuration:

{
  "mcpServers": {
    "pure.md": {
      "command": "npx",
      "args": ["-y", "puremd-mcp"],
      "env": {
        "PUREMD_API_KEY": "<TOKEN>"
      }
    }
  }
}

5. Save and restart Cursor.

Cline

No Cline-specific instructions are provided in the repository or documentation.

Securing API Keys (Environment Variables)

API keys should be stored securely using the env field in your MCP server configuration:

{
  "mcpServers": {
    "pure.md": {
      "command": "npx",
      "args": ["-y", "puremd-mcp"],
      "env": {
        "PUREMD_API_KEY": "<TOKEN>"
      }
    }
  }
}

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "MCP-name": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “MCP-name” to whatever the actual name of your MCP server is (e.g., “github-mcp”, “weather-api”, etc.) and replace the URL with your own MCP server URL.

Overview

Based on the available documentation, puremd-mcp is a focused and practical MCP server with clear utility for web unblocking and search. However, it lacks prompt templates and resource definitions, and there’s no mention of Roots or Sampling support. For general web access and search, it earns a solid score for simplicity and immediate usefulness.

MCP Score