Web Auth MCP Server
Enables executing HTTP requests with automatic browser-based authentication, supporting OAuth, form login, and cookie/session management.
README
Web Auth MCP Server
An MCP (Model Context Protocol) server that handles HTTP requests with automatic browser-based authentication.
Features
- Execute HTTP requests through MCP tools
- Automatic detection of authentication requirements
- Browser-based authentication for OAuth flows and login forms
- Chrome password manager integration - automatically use saved passwords
- Automatic form filling and submission - streamlined login process
- Retry requests with captured authentication credentials
- Support for cookies, tokens, and session management
- Configurable browser profiles (system or temporary)
Installation
Quick Install
./install.sh
Manual Install
# Create virtual environment
python3 -m venv venv
source venv/bin/activate
# Install dependencies
pip install -r requirements.txt
# Install package
pip install -e .
Usage
Running the MCP Server
Stdio Mode (Default)
# Activate virtual environment
source venv/bin/activate
# Run the server with stdio transport
web-auth-mcp
HTTP/SSE Mode
# Activate virtual environment
source venv/bin/activate
# Run the server with HTTP transport on default port 8080
web-auth-mcp-http
# Or specify custom host and port
web-auth-mcp --http 0.0.0.0 8080
# Access the server at: http://localhost:8080/sse
MCP Client Configuration
For Stdio Transport
Add to your MCP client configuration (e.g., Claude Desktop):
{
"mcpServers": {
"web-auth-mcp": {
"command": "python",
"args": ["-m", "web_auth_mcp.server"],
"env": {
"BROWSER_HEADLESS": "false",
"BROWSER_TIMEOUT": "60"
}
}
}
}
For HTTP/SSE Transport
Configure your MCP client to connect to the HTTP endpoint:
{
"mcpServers": {
"web-auth-mcp": {
"url": "http://localhost:8080/sse",
"env": {
"BROWSER_HEADLESS": "false",
"BROWSER_TIMEOUT": "60"
}
}
}
}
Testing
# Run tests
pytest
# Run example
python example_usage.py
Configuration
Create a .env file with optional configuration:
# Browser configuration
BROWSER_HEADLESS=false
BROWSER_TIMEOUT=60
BROWSER_WINDOW_SIZE=1920x1080
# Chrome Password Manager (NEW!)
BROWSER_USE_DEFAULT_PROFILE=false
BROWSER_ENABLE_PASSWORD_MANAGER=true
BROWSER_AUTO_FILL_PASSWORDS=true
# Authentication configuration
AUTH_CACHE_TTL=3600
AUTH_RETRY_ATTEMPTS=3
# Authentication timing (fixes quick closing issue)
BROWSER_TIMEOUT=300
MANUAL_AUTH_TIMEOUT=180
AUTH_WAIT_TIME=10
# Logging configuration
LOG_LEVEL=INFO
SUPPRESS_MCP_WARNINGS=true
Chrome Password Manager Integration
The server now supports using Chrome's saved passwords for automatic authentication:
Quick Start
-
Enable password manager features:
export BROWSER_ENABLE_PASSWORD_MANAGER=true export BROWSER_AUTO_FILL_PASSWORDS=true export BROWSER_HEADLESS=false -
Choose profile mode:
- System Profile (access to all saved passwords):
export BROWSER_USE_DEFAULT_PROFILE=true - Temporary Profile (session-only passwords):
export BROWSER_USE_DEFAULT_PROFILE=false
- System Profile (access to all saved passwords):
-
Test the functionality:
python test_password_manager.py python example_password_manager.py
How It Works
- Automatic Detection: Detects login forms on web pages
- Chrome Autofill: Triggers Chrome's built-in password autofill
- Form Submission: Automatically submits forms when credentials are found
- Fallback: Manual interaction available if auto-fill fails
Supported Login Forms
- Standard HTML login forms
- Email/username + password combinations
- Common form field patterns and selectors
- Various submit button types
For detailed documentation, see CHROME_PASSWORD_MANAGER.md.
Troubleshooting
MCP Initialization Warnings
You may see warnings like:
WARNING - Failed to validate request: Received request before initialization was complete
This is normal MCP protocol behavior and doesn't affect functionality. These warnings occur during the client-server handshake when requests are received before initialization completes.
To suppress these warnings:
export SUPPRESS_MCP_WARNINGS=true
# or add to your .env file:
echo "SUPPRESS_MCP_WARNINGS=true" >> .env
The server will continue to function correctly regardless of these warnings.
Tools
http_request
Execute HTTP requests with automatic authentication handling.
Parameters:
url(string): The URL to requestmethod(string): HTTP method (GET, POST, PUT, DELETE, etc.)headers(object, optional): Request headersbody(string, optional): Request bodyauth_required(boolean, optional): Force authentication flow
Returns:
- Response status, headers, and body
- Authentication details if authentication was performed
How It Works
- HTTP Request: The server receives an HTTP request through the MCP tool
- Authentication Detection: Checks if authentication is required by examining:
- HTTP status codes (401, 403)
- Redirect responses to login pages
- Content indicators ("please log in", etc.)
- WWW-Authenticate headers
- Browser Authentication: If authentication is needed:
- Opens a browser window (headless or visible)
- Navigates to the URL requiring authentication
- Waits for user to complete authentication
- Extracts authentication data (cookies, tokens, headers)
- Request Retry: Retries the original request with authentication data
- Response: Returns the final response with authentication status
Supported Authentication Types
- OAuth 2.0 flows (Authorization Code, Implicit, etc.)
- Form-based login (username/password forms)
- Cookie-based sessions
- Token-based authentication (JWT, Bearer tokens)
- CSRF protection (automatic token extraction)
Browser Automation
The server uses Selenium WebDriver with Chrome to handle authentication:
- Supports both headless and visible browser modes
- Automatically detects authentication completion
- Extracts tokens from localStorage, sessionStorage, and cookies
- Handles CSRF tokens from meta tags
- Configurable timeouts and window sizes
Troubleshooting
Browser Issues
Browser doesn't open:
- Check that
BROWSER_HEADLESS=false(default) - Ensure Chrome is installed on your system
Browser opens but doesn't close:
- The server detects authentication completion automatically
- Wait for the success message: "✅ Authentication completed successfully!"
- If stuck, check that you've completed the login process fully
Authentication timeout:
- Increase timeout:
BROWSER_TIMEOUT=120(default: 60 seconds) - Ensure you complete authentication within the timeout period
Common Solutions
# Force visible browser
export BROWSER_HEADLESS=false
# Increase timeout
export BROWSER_TIMEOUT=120
# Debug mode
export LOG_LEVEL=DEBUG
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.