Give Your AI Agent Security Superpowers with Intruder MCP

Stop switching between your IDE, Claude, and security dashboards. This MCP server connects your AI agents directly to Intruder's security scanning platform, turning conversational AI into a security automation powerhouse.

The Problem: Security Workflows Break Your Flow

You're deep in a code review when security questions arise. Instead of staying in your development environment, you're forced to:

• Switch to Intruder's web interface
• Manually trigger scans
• Wait for results, then context-switch back
• Repeat this dance for every security question

Your AI assistant knows your codebase intimately but can't help with the security analysis you actually need.

The Solution: Security Scanning Through Conversation

This MCP server bridges that gap. Your AI agents can now directly:

• Launch targeted scans on specific hosts or network ranges
• Retrieve scan results and analyze vulnerabilities in context
• Manage your asset inventory by adding new targets
• Generate security reports without leaving your conversation

Real-World Use Cases

Code Review Security Checks

"Can you scan staging.myapp.com and check if the new authentication 
endpoints have any obvious vulnerabilities?"

Your AI agent launches an Intruder scan, waits for results, and provides analysis directly in your code review tool.

Infrastructure Security Audits

"I've deployed these 5 new servers. Add them to our monitoring 
and run baseline scans on all of them."

The agent adds assets to Intruder, triggers scans, and reports back with a consolidated security assessment.

Incident Response

"We got reports of suspicious activity on api.example.com. 
What does our latest scan show for that host?"

Instead of logging into multiple dashboards, get immediate security intelligence through your AI assistant.

Continuous Security Integration

"Set up weekly enhanced scans for all production assets and 
alert me if any new high-severity issues are found."

Your AI agent becomes your security operations assistant, handling routine scanning and triage.

Integration Made Simple

The server works with any MCP-compatible client - Claude Desktop, Cursor, or custom implementations. Three deployment options:

Smithery (Fastest) One-click deployment through the Smithery marketplace. Just add your Intruder API key.

Local Development

uv venv && uv pip install -e .

Add to your MCP client config and you're running locally in under 2 minutes.

Docker Production

docker run --env INTRUDER_API_KEY=your-key ghcr.io/intruder-io/intruder-mcp

Production-ready container with your API credentials.

Why This Changes Everything

Security scanning typically requires dedicated tools and interfaces. This MCP server eliminates that friction by bringing enterprise security capabilities directly into your AI-powered development workflow.

You're not just saving time - you're enabling entirely new security workflows that weren't practical before. When your AI assistant can both understand your code and actively scan for vulnerabilities, security becomes a natural part of every development conversation rather than a separate, disruptive process.

Get started at smithery.ai/server/@intruder-io/intruder-mcp or clone the GitHub repository to run locally.