Skip to main content
Glama
webfetch-camouflage

webfetch-camouflage

by tianhuil

webfetch-camouflage-mcp

Tests License: MIT

alt text

Webfetch Camouflage is an MCP server for fetching web content with browser camouflage using curl_cffi. It mimics real browser TLS/HTTP2 fingerprints to bypass bot detection, returning clean Markdown output converted from HTML.

Installation

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "webfetch-camouflage": {
      "command": "uvx",
      "args": ["git+https://github.com/tianhuil/webfetch-camouflage-mcp.git"]
    }
  }
}
claude mcp add-json webfetch-camouflage '{"command":"uvx","args":["git+https://github.com/tianhuil/webfetch-camouflage-mcp.git"]}'
code --add-mcp '{"name":"webfetch-camouflage","command":"uvx","args":["git+https://github.com/tianhuil/webfetch-camouflage-mcp.git"]}'
gemini mcp add webfetch-camouflage uvx -- git+https://github.com/tianhuil/webfetch-camouflage-mcp.git
codex mcp add webfetch-camouflage -- uvx git+https://github.com/tianhuil/webfetch-camouflage-mcp.git

Add to opencode.json:

macOS/Linux: ~/.local/share/opencode/opencode.json Windows: %LOCALAPPDATA%\opencode\opencode.json

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "webfetch-camouflage": {
      "type": "local",
      "command": ["uvx", "git+https://github.com/tianhuil/webfetch-camouflage-mcp.git"],
      "enabled": true
    }
  }
}

Related MCP server: Fetch MCP

Tool reference

The fetch_url tool accepts:

  • url (required): URL to fetch

  • impersonate (optional, default "chrome"): browser profile to impersonate

  • timeout (optional, default 10): request timeout in seconds

  • max_chars (optional): maximum characters to return; truncates with ... if set

Supported impersonation profiles:

  • Chrome: chrome99, chrome100, chrome101, chrome104, chrome107, chrome110, chrome116, chrome119, chrome120, chrome123, chrome124, chrome131, chrome133a, chrome136

  • Firefox: firefox133, firefox135

  • Safari: safari153, safari155, safari170, safari180, safari184, safari260

  • Edge: edge99, edge101, edge133, edge135

Notes on Development

Setup

git clone https://github.com/tianhuil/webfetch-camouflage-mcp.git
cd webfetch-camouflage-mcp
uv sync

Running the server

uv run webfetch-camouflage-mcp

Available tasks

poe test       # Run tests
poe lint       # Lint code
poe format     # Format code
poe typecheck  # Type check
uv run pip-audit  # Security audit

MCP server testing

Test the server directly via JSON-RPC:

echo '{"jsonrpc": "2.0", "id": 1, "method": "initialize", "params": {"protocolVersion": "2024-11-05", "capabilities": {}, "clientInfo": {"name": "test", "version": "1.0"}}}' | uv run webfetch-camouflage-mcp

echo '{"jsonrpc": "2.0", "id": 1, "method": "initialize", "params": {"protocolVersion": "2024-11-05", "capabilities": {}, "clientInfo": {"name": "test", "version": "1.0"}}}' | uvx git+https://github.com/tianhuil/webfetch-camouflage-mcp.git

Or use the helper script:

uv run python src/get_tool_details.py

Available Tools

1 tool
fetch_urlA

Fetch web content from a URL with browser camouflage.

Args: url: The URL to fetch impersonate: Browser to impersonate (default: 'chrome'). Can be any string - curl_cffi will attempt to use the corresponding browser fingerprint. Common values include: chrome99, chrome100, chrome101, chrome104, chrome107, chrome110, chrome116, chrome119, chrome120, chrome123, chrome124, chrome131, chrome133a, chrome136, chrome99_android, chrome131_android, safari153, safari155, safari170, safari180, safari184, safari260, safari172_ios, safari180_ios, safari184_ios, safari260_ios, firefox133, firefox135, firefox135_android, tor145, edge99, edge101, edge133, edge135 timeout: Request timeout in seconds (default: 10) max_chars: Maximum number of characters to return in the response. If None, returns the full content (default: None)

Returns: The fetched content converted to Markdown format, or an error message if the request fails.

Inputs: url (required), impersonate, timeout, max_chars. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.4/5.0

    Scored across 1 tool

    Disambiguation5/5

    Only one tool exists, so there is no risk of confusion between tools.

    Naming Consistency5/5

    The single tool name 'fetch_url' follows a clear verb_noun pattern, consistent with best practices.

    Tool Count4/5

    With one tool, the server is minimal but still reasonable for its narrow purpose of fetching URLs with browser camouflage.

    Completeness5/5

    The tool fully covers the advertised functionality (fetching web content with impersonation), with no obvious gaps for its stated domain.

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      C
      maintenance
      An MCP server that retrieves web page content using Playwright headless browser, capable of extracting main content and converting to Markdown format.
      3
      1,088
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Remote MCP server for web scraping with anti-bot evasion. Provides stealth HTTP fetching, headless browser with Cloudflare bypass, CSS selectors, YouTube transcripts, and Markdown conversion.
      1
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      MCP server for retrieving web pages as clean Markdown, with configurable detail levels and optional Chromium rendering for JavaScript-heavy pages.
      1
      MIT