SAFE-MCP Security Hackathon
Hackathon Team please check in here:
Open Google Form
SAFE-MCP Security Hackathon
Help secure AI agent ecosystems by finding and documenting MCP vulnerabilities!
Schedule:
11:00 AM - SAFE-MCP project walkthrough & vulnerable server demo
11:30 AM - Hackathon begins!
12:00 PM - Pizza on the house 🍕
3:30 PM - Demo presentations
4:00 PM - Winner announcement
Two Ways to Participate:
🔧 Open Source Track: Add new attack techniques, mitigations, and detections to the SAFE-MCP project
💻 Proof-of-Concept Track: Build demos showing MCP vulnerabilities and their fixes
Who Should Join:
Security researchers, developers, students, and anyone curious about AI security
What You'll Get:
Templates and guidelines for easy participation
Vulnerable MCP server for testing and exploitation
SAFE-MCP core team guidance to help create your PR
Real impact on AI security standards
🚀 Come Prepared:
Please have these ready before the event:
Fork and pull the SAFE-MCP repository
Sample MCP servers for testing and exploitation
Basic Git setup for commits
All skill levels welcome! Come help build a safer future for AI agents.
Discord: https://discord.com/invite/n6tS37uQ
----------------------------------------------
Contribute to SAFE-Model Context Protocol
-----------------------------------------------
SAFE-MCP specifically tackles critical security concerns arising from AI agents interacting with external tools, data sources, and workflows.
Key MCP Security Challenges addressed by SAFE-MCP:
Prompt Injection and Unauthorized Tool Execution
Excessive OAuth Permissions and Token Management
Third-party MCP Server Risks
Supply-Chain Vulnerabilities and Tool Poisoning
Frederick Kautz, a recognized authority in cloud-native security, leads this initiative. His distinguished background includes:
Co-authoring NIST Special Publication 800-204D, foundational in software supply chain security.
Co-founding Network Service Mesh, transforming secure cloud-native networking.
Leadership in KubeCon + CloudNativeCon and Zero Trust architectures.
SPIFFE Steering Committee Member
Join Frederick and industry experts to explore how SAFE-MCP sets new standards for secure MCPs.
Github: fkautz/safe-mcp
Connect with the Community:
Join our AI Agents Community Discord: https://discord.gg/n6tS37uQ
Learn more: SAFE-MCP GitHub Repository
Throughout the Event: Enjoy pizza and drinks on the house.
Judging Panel
Frederick Kautz: Distinguished leader in AI, zero-trust, open-source and cloud-native.
Arjun Subedi: Co-founder Astha.ai
Saleh Alothaim: Angel Investor
Bishnu Bista: Core contributor SAFE-MCP