🔒 Security & Trust
Our commitment to keeping MCPLookup.org secure and trustworthy for the entire AI ecosystem
Security Overview
Security is fundamental to MCPLookup.org's mission of providing a trusted discovery service for Model Context Protocol servers. We implement multiple layers of security to protect our users and the integrity of the MCP ecosystem.
Encryption
All data encrypted in transit
Verification
DNS-based domain ownership
Monitoring
24/7 security monitoring
Infrastructure Security
Serverless Architecture
Our serverless design eliminates many traditional attack vectors by having no persistent servers to compromise. Each request is handled by isolated functions that automatically scale and terminate.
Minimal Data Storage
We use Upstash Redis (serverless) for registered server metadata only. All data has automatic TTL expiration. No traditional databases, no file system storage, minimal attack surface.
Edge Network
Deployed on Vercel's global edge network with built-in DDoS protection, automatic SSL/TLS certificates, and geographic distribution for resilience.
Domain Verification
We use cryptographic DNS verification to ensure only legitimate domain owners can register MCP servers:
Verification Process:
1. Generate unique cryptographic challenge for each registration
2. Require TXT record addition to domain's DNS
3. Verify record exists and matches expected value
4. Only then allow server to be discoverable
Security Benefit: This prevents unauthorized users from registering servers for domains they don't control, maintaining trust in the discovery service.