Overview
A Model Context Protocol (MCP) server that provides advanced HTTP request capabilities for Claude and other LLMs. Built on the rquest library, it enables realistic browser emulation with accurate TLS, JA3, and JA4 fingerprints, allowing models to interact with websites more naturally and bypass common anti-bot measures. It also supports converting PDF and HTML documents to Markdown to streamline processing by LLMs. Features include complete HTTP methods, authentication support, and extensive request customization. The server handles large responses with token counting, converts HTML and PDFs to Markdown for easier processing, and securely stores responses in system temporary directories. It exposes tools to retrieve stored responses and convert them to Markdown, aiding model workflows. Security-focused TLS is emphasized via BoringSSL and realistic fingerprints, with browser profiles such as Chrome, Firefox, Safari, Edge, and OkHttp. PDF support includes automatic content-type detection and pre-downloaded models for performance. Installation options include uv for development and pip for production, with configuration paths for Claude.app integrations.
Features
Complete HTTP Methods
Supports GET, POST, PUT, DELETE, PATCH, HEAD, OPTIONS, and TRACE for comprehensive HTTP interactions.
Browser Fingerprinting
Accurate TLS, JA3/JA4, and HTTP/2 browser fingerprints to emulate real browsers.
Content Handling
Automatic handling of large responses with token counting and secure storage.
HTML to Markdown Conversion
Converts HTML responses to Markdown for easier processing by LLMs.
PDF to Markdown Conversion
Converts PDFs to Markdown using the Marker library for high-quality output.
Authentication Support
Supports Basic, Bearer, and custom authentication methods.
Request Customization
Supports headers, cookies, redirects, form data, JSON payloads, multipart, and query parameters.
SSL Security
Uses BoringSSL for secure connections with realistic browser fingerprints.
Who Is This For?
- LLM developers:Enable realistic browser-like HTTP requests for Claude and other LLMs.
- Web data engineers:Fetch, process, and convert HTML/PDF content for ML workflows in stable pipelines.
