MCP.so
Sign In

mcp-rquest

@xxxbrian

About mcp-rquest

A MCP server providing realistic browser-like HTTP request capabilities with accurate TLS/JA3/JA4 fingerprints for bypassing anti-bot measures. It also supports converting PDF and HTML documents to Markdown for easier processing by LLMs.

Basic information

Category

Files & Storage

License

MIT

Runtime

python

Transports

stdio

Publisher

xxxbrian

Submitted by

Bojin Li

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "http-rquest": {
      "command": "uvx",
      "args": [
        "mcp-rquest"
      ]
    }
  }
}

Tools

12

Perform GET requests with optional parameters

Submit data via POST requests

Update resources with PUT requests

Remove resources with DELETE requests

Partially update resources

Retrieve only headers from a resource

Retrieve options for a resource

Diagnostic request tracing

Retrieve stored large responses, optionally by line range

Convert HTML or PDF responses to Markdown format for better LLM processing

Get the current state of the PDF models loading process

Restart the PDF models loading process if it failed or got stuck

Overview

What is mCP-rquest?

mcp-rquest is an MCP server that provides advanced HTTP request capabilities with realistic browser fingerprinting, allowing LLMs like Claude to interact with websites more naturally and bypass common anti-bot measures. It also converts HTML and PDF documents to Markdown for easier LLM processing.

How to use mcp-rquest?

Install using uvx (recommended) or pip, then configure the server in your MCP client (e.g., Claude for Desktop) using the provided JSON settings. Once configured, you can invoke HTTP request tools such as http_get or http_post, and use response handling tools like get_stored_response_with_markdown to convert responses to Markdown.

Key features of mcp-rquest

  • Complete HTTP methods: GET, POST, PUT, DELETE, PATCH, HEAD, OPTIONS, TRACE
  • Realistic browser fingerprinting with TLS, JA3/JA4 fingerprints
  • HTML and PDF to Markdown conversion for LLM processing
  • Automatic handling of large responses with token counting
  • Authentication support: Basic, Bearer, and custom methods
  • SSL security using BoringSSL with browser-like fingerprints

Use cases of mcp-rquest

  • Bypass anti-bot measures to access web content programmatically
  • Convert web pages and PDFs to Markdown for LLM consumption
  • Perform authenticated API requests with realistic browser fingerprints
  • Retrieve and store large HTTP responses for later analysis
  • Test website options, headers, and routes using various HTTP methods

FAQ from mcp-rquest

What browsers can mcp-rquest emulate?

It supports Chrome, Firefox, Safari (including iOS and iPad), Edge, and OkHttp.

Does mcp-rquest support PDF conversion?

Yes, it uses the Marker library to convert PDF files to Markdown. Models are pre-downloaded during installation.

How do I install mcp-rquest?

Use uvx mcp-rquest (recommended) or install via pip with pip install mcp-rquest and run with python -m mcp_rquest.

What authentication methods are supported?

It supports Basic, Bearer, and custom authentication methods.

Is there any data stored locally?

Yes, large responses are stored securely in a system temporary directory for later retrieval via tools like get_stored_response.

Comments

More Files & Storage MCP servers