Which integrations are available for this server?

Provides capabilities for bypassing Cloudflare Turnstile captchas and interacting with Cloudflare-protected sites.

How do I use PyPen MCP?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@PyPen MCP Capture network requests from the login page and check for security issues" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

PyPen MCP

by eyenoticeall

Overview Schema Source Related Servers Discussions Admin

Python

Local

PyPen MCP

Model Context Protocol server for browser-based penetration testing using PyDoll.

Overview

PyPen MCP provides a comprehensive set of tools for LLM-powered browser automation and security testing. It leverages PyDoll's Chrome DevTools Protocol integration for reliable, webdriver-free browser control.

Related MCP server: nodriver-proxy-mcp

Features

Browser Management

Launch/close browser instances with configurable options
Navigation control (navigate, back, forward, refresh)
Headless mode, proxy support, custom user agents
Incognito mode and resource blocking

Network Operations

Monitoring: Capture all HTTP traffic, analyze requests/responses
Interception: Block, modify, or mock requests in real-time
Authenticated Requests: Make API calls that inherit browser session/cookies

DOM Manipulation

Find elements by CSS selector, XPath, or attributes
Click, type, scroll, and interact with elements
Extract page source, element text, and HTML
Human-like scrolling with physics simulation

JavaScript Execution

Execute arbitrary JavaScript in browser context
Access localStorage and sessionStorage
Extract global variables, forms, and links

Session Management

Cookie manipulation (get, set, delete, clear)
Import/export cookies in multiple formats
Session-aware HTTP requests

Captcha Handling

Turnstile Bypass: Automatic Cloudflare Turnstile interaction
Human Intervention: Request human help for unsolvable captchas (ReCAPTCHA v2, hCaptcha)
Detection: Identify captcha types on the page

Debugging

Screenshots (full page or elements)
Viewport control
Performance metrics
Element highlighting

Installation

pip install pypen-mcp

Or install from source:

git clone https://github.com/your-repo/pypen-mcp
cd pypen-mcp
pip install -e .

Configuration

Add to your MCP client configuration:

{
  "mcpServers": {
    "pypen": {
      "command": "pypen-mcp"
    }
  }
}

For development:

{
  "mcpServers": {
    "pypen": {
      "command": "/path/to/pypen-mcp/.venv/bin/pypen-mcp"
    }
  }
}

Usage Examples

Basic Browser Automation

# Launch browser
browser_launch(headless=True)

# Navigate to target
browser_navigate(url="https://example.com")

# Extract page content
dom_get_source()

# Find and interact with elements
dom_find_element(css_selector="#username")
dom_type(selector="#username", text="admin")
dom_click(selector="button[type=submit]")

# Take screenshot for documentation
debug_screenshot(path="/tmp/screenshot.png")

# Close when done
browser_close()

Network Monitoring

# Enable traffic capture
network_enable_monitoring()

# Navigate and capture
browser_navigate(url="https://api.example.com")

# Get captured requests
network_get_logs(filter_url="/api/")

# Get response body for specific request
network_get_response_body(request_id="abc123")

# Disable when done
network_disable_monitoring()

Request Interception

# Set up interception rules
network_setup_handler(
    block_patterns=["analytics", "tracking", "ads"],
    modify_headers={"Authorization": "Bearer token123"},
    mock_responses={"/api/config": {"status": 200, "body": {"debug": true}}}
)

# Navigate with interception active
browser_navigate(url="https://example.com")

Session-Based Requests

# Login via UI (handles complex auth)
browser_navigate(url="https://app.com/login")
dom_type(selector="#email", text="user@example.com")
dom_type(selector="#password", text="password123")
dom_click(selector="button[type=submit]")

# Make authenticated API calls
session_make_request(
    url="https://app.com/api/user/profile",
    method="GET"
)

Captcha Handling

# Enable automatic Turnstile bypass
captcha_enable_turnstile_bypass(time_to_wait_captcha=10.0)

# Navigate to protected site
browser_navigate(url="https://protected-site.com")

# Wait for captcha processing
# ... automation continues ...

# For unsolvable captchas, request human help
captcha_detect_type()  # Check what's on page
captcha_request_human_intervention(
    captcha_type="recaptcha_v2",
    message="Please solve the ReCAPTCHA to continue"
)

# Wait for human to solve
captcha_wait_for_resolution(request_id="abc123", timeout=300)

Tool Reference

Browser Tools

Tool	Description
`browser_launch`	Launch browser with options (headless, proxy, etc.)
`browser_close`	Close the browser instance
`browser_navigate`	Navigate to a URL
`browser_go_back`	Navigate back in history
`browser_go_forward`	Navigate forward in history
`browser_refresh`	Refresh current page
`browser_get_info`	Get current URL and title

Network Tools

Tool	Description
`network_enable_monitoring`	Start capturing HTTP traffic
`network_disable_monitoring`	Stop traffic capture
`network_get_logs`	Get captured requests (with optional filter)
`network_get_response_body`	Get response body for a request
`network_enable_interception`	Enable request interception
`network_disable_interception`	Disable interception
`network_setup_handler`	Configure auto interception rules

DOM Tools

Tool	Description
`dom_find_element`	Find element by selector
`dom_find_elements`	Find multiple elements
`dom_get_text`	Get element text content
`dom_get_html`	Get element HTML
`dom_click`	Click an element
`dom_type`	Type text into input
`dom_scroll`	Scroll the page
`dom_get_source`	Get full page HTML
`dom_wait_for`	Wait for element to appear

JavaScript Tools

Tool	Description
`js_execute`	Execute JavaScript code
`js_get_console_logs`	Get console logs
`js_get_global_vars`	Extract global variables
`js_get_local_storage`	Get localStorage data
`js_get_session_storage`	Get sessionStorage data
`js_set_local_storage`	Set localStorage value
`js_get_forms`	Get all forms with inputs
`js_get_links`	Get all links on page

Session Tools

Tool	Description
`session_get_cookies`	Get all cookies
`session_get_cookie`	Get specific cookie
`session_set_cookie`	Set a cookie
`session_delete_cookie`	Delete a cookie
`session_clear_cookies`	Clear all cookies
`session_export_cookies`	Export cookies (json/netscape/header)
`session_import_cookies`	Import cookies from JSON
`session_make_request`	Make session-authenticated HTTP request

Captcha Tools

Tool	Description
`captcha_enable_turnstile_bypass`	Enable auto Turnstile handling
`captcha_disable_turnstile_bypass`	Disable Turnstile handling
`captcha_detect_type`	Detect captcha type on page
`captcha_handle_auto`	Auto-handle detected captcha
`captcha_request_human_intervention`	Request human help
`captcha_get_pending_interventions`	Get unresolved requests
`captcha_resolve_intervention`	Mark intervention as resolved
`captcha_wait_for_resolution`	Wait for human to solve

Debug Tools

Tool	Description
`debug_screenshot`	Take page screenshot
`debug_get_viewport`	Get viewport size
`debug_set_viewport`	Set viewport size
`debug_get_performance`	Get performance metrics
`debug_highlight`	Highlight matching elements

Important Notes

Browser Session Persistence

The browser session persists until explicitly closed with browser_close
All operations (DOM, JS, network) use the same active session
Cookies and session state are maintained across navigations

Captcha Success Factors

Turnstile bypass success depends on:

IP Reputation: Use residential proxies with good reputation
Browser Fingerprint: Configure realistic browser options
Behavioral Patterns: Add realistic delays and interactions

Human Intervention Workflow

Detect captcha: captcha_detect_type()
Try auto-handling: captcha_handle_auto()
If fails, request help: captcha_request_human_intervention()
Wait for resolution: captcha_wait_for_resolution()
Continue automation

License

MIT

Available Tools

52 tools

browser_closeA

Close the browser instance

Inputs: no named parameters.

browser_get_infoA

Get information about the current page (URL, title)

Inputs: no named parameters.

browser_go_backA

Navigate back in browser history

Inputs: no named parameters.

browser_go_forwardA

Navigate forward in browser history

Inputs: no named parameters.

browser_launchB

Launch a new browser instance with optional configuration (headless, proxy, user agent, etc.)

Inputs: headless, proxy, user_agent, binary_location, +4 more.

browser_navigateB

Navigate to a URL

Inputs: url (required).

browser_refreshA

Refresh the current page

Inputs: no named parameters.

captcha_detect_typeB

Detect what type of captcha is present on the current page. Scans for Cloudflare Turnstile, reCAPTCHA, hCaptcha, and image challenge indicators.

Inputs: no named parameters.

captcha_disable_turnstile_bypassA

Disable automatic Turnstile bypass

Inputs: no named parameters.

captcha_enable_turnstile_bypassA

Enable automatic Cloudflare Turnstile bypass. Uses PyDoll's built-in mechanism to detect and click the Turnstile checkbox. Success depends on IP reputation and browser fingerprint - this is NOT a magic bypass. Use residential proxies for best results.

Inputs: time_to_wait_captcha.

captcha_get_pending_interventionsA

Get all pending human intervention requests that have not been resolved

Inputs: no named parameters.

captcha_handle_autoA

Attempt to automatically handle any detected captcha. For Turnstile: uses built-in bypass. For others (ReCAPTCHA v2, hCaptcha, image challenges): returns need for human intervention.

Inputs: timeout.

captcha_request_human_interventionA

Request human intervention for an unsolvable captcha (e.g., ReCAPTCHA v2 image challenges, hCaptcha puzzles). Creates an intervention request with screenshot. The automation should wait for human to solve.

Inputs: captcha_type (required), message (required), take_screenshot.

captcha_resolve_interventionA

Mark an intervention request as resolved after human has solved the captcha. This allows the automation to continue.

Inputs: request_id (required), resolution.

captcha_wait_for_resolutionA

Block until a human intervention is resolved or timeout expires. Use this to pause automation while waiting for human to solve captcha.

Inputs: request_id (required), timeout, poll_interval.

debug_get_performanceB

Get performance metrics from the page

Inputs: no named parameters.

debug_get_viewportA

Get the current viewport size

Inputs: no named parameters.

debug_highlightB

Highlight all elements matching a selector

Inputs: selector (required), color.

debug_screenshotA

Take a screenshot of the current page. Returns the image directly. MUST be called after every dom_click to observe the result.

Inputs: no named parameters.

debug_set_viewportB

Set the viewport size

Inputs: width (required), height (required).

dom_clickB

Click on an element. IMPORTANT: You SHOULD either call debug_screenshot or view the source after every click to observe the result of the interaction.

Inputs: selector (required), timeout.

dom_find_elementC

Find an element on the page using various selectors

Inputs: tag_name, id, class_name, name, +4 more.

dom_find_elementsC

Find multiple elements on the page

Inputs: tag_name, class_name, css_selector, xpath, +1 more.

dom_get_htmlB

Get the outer HTML of an element

Inputs: selector (required), timeout.

dom_get_sourceB

Get the full page HTML source

Inputs: no named parameters.

dom_get_textB

Get the text content of an element

Inputs: selector (required), timeout.

dom_scrollB

Scroll the page

Inputs: direction, amount.

dom_typeC

Type text into an input element

Inputs: selector (required), text (required), clear_first, timeout.

dom_wait_forB

Wait for an element to appear on the page

Inputs: selector (required), timeout.

js_executeB

Execute JavaScript in the browser context

Inputs: script (required).

js_get_console_logsB

Get console logs from the browser

Inputs: no named parameters.

js_get_formsA

Get all forms on the page with their inputs

Inputs: no named parameters.

js_get_global_varsA

Extract global JavaScript variables

Inputs: no named parameters.

js_get_linksA

Get all links on the page

Inputs: no named parameters.

js_get_local_storageA

Get all localStorage data

Inputs: no named parameters.

js_get_session_storageA

Get all sessionStorage data

Inputs: no named parameters.

js_set_local_storageC

Set a localStorage value

Inputs: key (required), value (required).

network_disable_interceptionB

Disable request interception

Inputs: no named parameters.

network_disable_monitoringB

Disable network event monitoring

Inputs: no named parameters.

network_enable_interceptionB

Enable request/response interception

Inputs: resource_type, request_stage.

network_enable_monitoringB

Enable network event monitoring to capture all HTTP traffic

Inputs: no named parameters.

network_get_logsA

Get captured network logs, optionally filtered by URL pattern

Inputs: filter_url.

network_get_response_bodyB

Get the response body for a specific request ID

Inputs: request_id (required).

network_setup_handlerB

Set up automatic interception handler with blocking, modification, and mocking rules

Inputs: block_patterns, modify_headers, mock_responses.

session_clear_cookiesC

Clear all cookies

Inputs: no named parameters.

session_delete_cookieB

Delete a cookie by name

Inputs: name (required), domain.

session_export_cookiesB

Export cookies in various formats (json, netscape, header)

Inputs: format.

session_get_cookieB

Get a specific cookie by name

Inputs: name (required).

session_get_cookiesB

Get all cookies from the browser

Inputs: no named parameters.

session_import_cookiesB

Import cookies from JSON string

Inputs: cookies_json (required).

session_make_requestC

Make an HTTP request using the browser's session (inherits cookies/auth)

Inputs: url (required), method, headers, body, +1 more.

session_set_cookieC

Set a cookie in the browser

Inputs: name (required), value (required), domain, path, +2 more.

Tool Schema Changelog

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

Loading tool schema changelog…

Deploy Server

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.6/5.0

Scored 2026-05-28 11:55 across 52 tools

Disambiguation5/5

Each tool has a clear, distinct purpose, with prefixes like browser_, captcha_, debug_, dom_, js_, network_, session_ grouping related functionality. No two tools perform the same action, and even within groups, tools are well-differentiated.

Naming Consistency5/5

All tool names follow a consistent snake_case convention with a prefix indicating the functional domain. This makes it easy for an agent to predict tool names and understand their roles.

Tool Count4/5

52 tools is on the high side, but the server covers a broad scope of browser automation features (browser control, DOM, JS, network, sessions, captchas). The count is appropriate for the comprehensive capability set, though slightly above the ideal range.

Completeness5/5

The tool surface is remarkably complete: browser lifecycle, navigation, DOM interactions, debugging, JavaScript execution, network monitoring/interception, cookie/session management, and captcha handling all have dedicated tools. No obvious gaps for the intended domain.

Resources

Related MCP Connectors

AdsCrawl
io.github.AdsCrawl
Cloud browser, web extraction, screenshots, and remote CDP tools for AI agents.
Browser Automation Web Scraping Agent Orchestration
1
https://api.adscrawl.net/mcp
browser-useA
com.browser-use
Browser agents and stealth cloud browsers for AI: run web tasks, fill forms, extract data.
Agent Orchestration Browser Automation Web Scraping
7170,259
https://api.browser-use.com/v3/mcp
Scrappy Cloud Browser
gg.scrappy
Cloud browsers for AI agents: saved logins, residential IPs, live human takeover, 57 browser tools.
Agent Orchestration Browser Automation Web Scraping
1
https://scrappy.gg/api/mcp
browser-forestA
com.browserforest
Undetectable cloud browser sessions for AI agents and scrapers. Navigate, extract, click, captcha.
Browser Automation Web Scraping Agent Orchestration
91
https://browserforest.com/api/mcp/bf

Related MCP Servers

MCP-Server-Playwright
Browser Automation Web Scraping
VikashLoomba
A
license
B
quality
D
maintenance
Enables LLMs to interact with web pages, take screenshots, and execute JavaScript in a real browser environment
Updated a year ago (2025-06-05 21:48 UTC)
10
264 npm
297
MIT
nodriver-proxy-mcp
Penetration Testing Browser Automation Code Execution
BobongKu
A
license
A
quality
D
maintenance
Gives AI agents a browser that bypasses bot detection, a MITM proxy for traffic interception, and a Python sandbox to autonomously write and execute security exploits.
Updated 7 months ago (2026-04-10 04:07 UTC)
39
3
MIT
Browser-Debugger
Testing & QA Tools Browser Automation
Selvadinesh-giga
F
license
Not graded
quality
D
maintenance
Enables AI agents to automate and debug real Chromium browsers with capabilities like screenshots, video recording, performance analysis, visual regression testing, and OCR text extraction.
Updated 9 months ago (2026-01-30 08:16 UTC)
13
-
browser-control-mcp-server
Browser Automation Web Scraping
yogesh-joshi-0333
A
license
A
quality
C
maintenance
Enables AI agents to fully control a browser for web automation, including navigation, clicking, typing, scrolling, screenshots, and DOM inspection, with session persistence and anti-bot bypass.
Updated 2 months ago (2026-09-07 11:19 UTC)
40
43 npm
MIT

PyPen MCP

Overview

Features

Browser Management

Network Operations

DOM Manipulation

JavaScript Execution

Session Management

Captcha Handling

Debugging

Installation

Configuration

Usage Examples

Basic Browser Automation

Network Monitoring

Request Interception

Session-Based Requests

Captcha Handling

Tool Reference

Browser Tools

Network Tools

DOM Tools

JavaScript Tools

Session Tools

Captcha Tools

Debug Tools

Important Notes

Browser Session Persistence

Captcha Success Factors

Human Intervention Workflow

License

Available Tools

Tool Schema Changelog

TDQS

Resources

Related MCP Connectors

Related MCP Servers

MCP-Server-Playwright

nodriver-proxy-mcp

Browser-Debugger

browser-control-mcp-server