About wass-mcp
WASS-MCP is an MCP server for authorized web application security testing. It combines several scanners, offers parallel full scans, and keeps execution history for later review.
Key features
- 01Nikto web server scanning
- 02Nuclei template-based vulnerability scans
- 03Wapiti web application scans
- 04Shcheck security header analysis
- 05Parallel scans with unified results
- 06Persistent scan execution history
Use cases
- 01Scan an authorized web server for vulnerabilities
- 02Review a site's HTTP security headers
- 03Run several scanners and inspect a combined report
- 04Browse or manage previous scan results
MCP support
WASS-MCP exposes scanning and history tools to MCP clients over Streamable HTTP, including individual scanner tools and a parallel full_scan tool.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEWASS-MCP
A Model Context Protocol (MCP) server for web application security scanning.
Features
- MCP Protocol Support - Full compatibility with MCP clients (Claude, etc.)
- Nikto Integration - Web server vulnerability scanning
- Nuclei Integration - Template-based vulnerability scanning
- Wapiti Integration - Web application vulnerability scanning
- Shcheck Integration - Security headers analysis
- Execution History - Persistent storage of scan results
- Stateless Design - Survives server restarts without session errors
- RESTful HTTP Transport - Streamable HTTP-based MCP protocol
Usage
docker run -p 127.0.0.1:8989:8989 tb0hdan/wass-mcp
MCP Client Configuration
Example command to add WASS-MCP server to Claude MCP clients:
claude mcp add wass-mcp --transport http http://127.0.0.1:8989
or Gemini:
gemini mcp add wass-mcp --transport http http://127.0.0.1:8989
Available Tools
nikto
Perform web server vulnerability scans using Nikto.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
host | string | Yes | Target hostname or IP address |
port | integer | No | Target port (default: 80) |
vhost | string | No | Virtual host header |
max_lines | integer | No | Maximum output lines |
offset | integer | No | Output line offset |
Example:
{
"host": "192.168.1.100",
"port": 443
}
nuclei
Perform template-based vulnerability scanning using Nuclei.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
host | string | Yes | Target hostname or IP address |
port | integer | No | Target port (default: 80) |
vhost | string | No | Virtual host header |
max_lines | integer | No | Maximum output lines |
offset | integer | No | Output line offset |
Vulnerabilities Detected:
- CVE detection via community templates
- Misconfigurations
- Exposed panels/dashboards
- Default credentials
- Technology detection
- Security headers analysis
- And many more via 8000+ community templates
Example:
{
"host": "192.168.1.100",
"port": 443
}
wapiti
Perform comprehensive web application vulnerability scans using Wapiti.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
host | string | Yes | Target hostname or IP address |
port | integer | No | Target port (default: 80) |
vhost | string | No | Virtual host header |
max_lines | integer | No | Maximum output lines |
offset | integer | No | Output line offset |
Vulnerabilities Detected:
- SQL Injection / Blind SQL Injection
- Cross-Site Scripting (XSS)
- File Inclusion / Path Traversal
- Command Execution
- CRLF Injection
- Server-Side Request Forgery (SSRF)
- Open Redirects
- HTTP Security Headers
- Content Security Policy issues
Example:
{
"host": "192.168.1.100",
"port": 8080
}
shcheck_py
Analyze HTTP security headers using shcheck.py.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
host | string | Yes | Target hostname or IP address |
port | integer | No | Target port (default: 80) |
vhost | string | No | Virtual host header |
max_lines | integer | No | Maximum output lines |
offset | integer | No | Output line offset |
Headers Analyzed:
- Content-Security-Policy
- Strict-Transport-Security
- X-Frame-Options
- X-Content-Type-Options
- Referrer-Policy
- Permissions-Policy
- And other security-relevant headers
Example:
{
"host": "example.com",
"port": 443
}
full_scan
Perform a comprehensive security scan using all available scanners in parallel.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
host | string | Yes | Target hostname or IP address |
port | integer | No | Target port (default: 80) |
vhost | string | No | Virtual host header |
max_lines | integer | No | Maximum output lines |
offset | integer | No | Output line offset |
Features:
- Runs nikto, nuclei, wapiti and shcheck scanners in parallel
- Merges results into a unified report
- Includes timing and status for each scanner
- Gracefully handles missing scanner binaries
Example:
{
"host": "192.168.1.100",
"port": 8080
}
history
Browse and manage tool execution history.
Parameters:
| Name | Type | Required | Description |
|---|---|---|---|
action | string | Yes | One of: list, get, delete, clear |
id | integer | For get/delete | Execution ID |
limit | integer | No | Results per page (default: 10) |
offset | integer | No | Pagination offset |
Actions:
list- List execution history with paginationget- Get full details of a specific executiondelete- Delete a specific execution by IDclear- Delete all execution history
API Endpoints
| Endpoint | Description |
|---|---|
POST /mcp | MCP protocol endpoint |
GET / | Service information (JSON) |
GET /debug/pprof/* | Profiling endpoints |
Development and advanced usage
Source build requirements
- Go 1.25+
- Nikto (
apt install niktoor equivalent) - Nuclei (
go install github.com/projectdiscovery/nuclei/v3/cmd/nuclei@latest) - Wapiti (
apt install wapitior equivalent) - Shcheck (
pip install shcheckor from GitHub) - SQLite3
# Clone the repository
git clone https://github.com/tb0hdan/wass-mcp.git
cd wass-mcp
# Build
make build
# Run
./build/wass-mcp
Starting the Server
# Default (localhost:8989)
./build/wass-mcp
# Custom bind address
./build/wass-mcp --bind 0.0.0.0:8080
# Custom database path
./build/wass-mcp --db /var/lib/wass-mcp/data.db
# Debug mode
./build/wass-mcp --debug
Configuration Options
| Flag | Default | Description |
|---|---|---|
--bind | localhost:8989 | HTTP server bind address |
--db | ./wass-mcp.db | SQLite database file path |
--debug | false | Enable debug logging |
--version | - | Print version and exit |
Linting
make lint
Testing
make test
Project Structure
wass-mcp/
├── cmd/wass-mcp/ # Application entry point
├── pkg/
│ ├── server/ # MCP server wrapper
│ ├── storage/ # Database layer (SQLite/GORM)
│ ├── models/ # Data models
│ ├── tools/ # MCP tool implementations
│ │ ├── nikto/ # Nikto web server scanner
│ │ ├── wapiti/ # Wapiti web app scanner
│ │ ├── nuclei/ # Nuclei template scanner
│ │ ├── shcheck/ # Security headers checker
│ │ ├── fullscan/ # Parallel full scan
│ │ └── history/ # History management
│ └── types/ # Shared types and constants
├── docs/ # Documentation
└── build/ # Build output and coverage reports
Security Notice
This tool is intended for authorized security testing only. Ensure you have proper authorization before scanning any systems. Unauthorized scanning may be illegal in your jurisdiction.
Project notes
For complete project notes, design decisions, and architecture overview, please refer to the Project Notes document.
License
BSD 3-Clause License - Copyright (c) 2026, Bohdan Turkynevych. See LICENSE for details.
Contributing
- Fork the repository
- Create a feature branch (
git checkout -b feature/new-tool) - Commit your changes (
git commit -am 'Add new scanning tool') - Push to the branch (
git push origin feature/new-tool) - Create a Pull Request
Acknowledgments
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.